Skip to main content

Full text of "Screen Volume 29 Issue 1"

See other formats

Volume 29 Number 1 Winter 1988 £4.50 $9.50 



This issue opens with a two-part study of the crises 
in ownership and ontology created by the new 
information technologies. John Frow focuses on the 
bizarre encounter of copyright law and computer 
software, while Bill Nichols pursues broader ques- 
tions of subjectivity and cybernetic systems. Their 
work returns Screen to concerns which grew out of 
the study of authorship in the ’70s, notably the 
attempt to develop a Marxist theory of copyright 
legislation and its juridical subject. 

In his Ownership of the Image ', Bernard Edelman 
examined the changing definitions of both the 
photographer and the photographed in nineteenth 
and twentieth century French copyright law. Until 
the early decades of this century, French law denied 
an author’s right over both photography and cinema 
on the grounds of its mechanical, ‘non-creative’ 
nature. But with the increase in capital investment 
in the cinema, Edelman argued, the French courts 
were obliged to reduce commercial risk by assigning 
exclusive rights to the product. The consequence 
was a redefinition of the cinematic process as a cre- 
ative one - a recording of an idea in sound and 
image - with the right of ownership vested first in 
the producer (as the financially responsible party) 
and, after subsequent struggles, ‘moral’ rights to 
recompense and reputation parcelled out among the 
creative personnel. ^ 

Historically, as Vincent Porter^ has reminded 
Edelman’s exponents, Anglo-American law has 
differed from its continental counterparts, originally 
defining copyright as a right to copy, a publisher’s 
right, rather than that of an author. But this has not 
exempted it from the contradictions posed by suc- 
cessive technological developments, which as John 
Frow observes, complicate distinctions between 
‘work’ and ‘copy’ (the player piano), between orig- 
inal research and ‘mere’ compilation (Dun and 
Bradstreet) and between receiver and transmitter 
(the VCR). Electronic information systems intensify 

these complications by their lack of material fixity, 
their blurring of human and artificial agency ^ and 
their commodification of knowledge itself - hitherto 
protected by a legal tradition which attempted to 
balance an individual right of ownership against a 
public right to the free circulation of ideas. 

Cybernetics, argues Bill Nichols, replaces mech- 
anical reproduction with electronic simulation, 
object with process. In the characteristic ontology of 
postmodernity, the biological mother of a child is 
designated a surrogate for its would-be adoptive 
parent and weapon systems are styled to resemble 
video games. Human intelligence is mechanised via 
computer metaphor, but the resulting ‘cyborg’ is 
then re-admitted to the realm of the organic. 

Both Frow and Nichols base their studies on cur- 
rent US copyright law. Meanwhile, a massive 
measure revising British legislation in this area, the 
Copyright, Designs and Patents Bill, is currently 
before Parliament. As this issue goes to press, the 
media education lobby is hopeful that the final Act 
will include three significant changes in copyright 

1) the extension of fair dealing in regard to the right 
of quotation from the mass media for the purposes 
of private study or broadcast criticism, review or 

* Bernard Edelman, Ownership of the Image, London, 
Routledge and Kegan Paul, 1979. See also, Paul Hirst and 
Elizabeth Kingdom, ‘On Edelman’s “Ownership of the 
Image”’, 5irezfi Winter 1979-80, vol 20 nos 3-4, pp 135-140. 

^ Vincent Porter, ‘Film Copyright and Edelman’s Theory of 
Law’, icreen Winter 1979-80, vol 20 nos 3-4, pp 14 1- 147. 
And, in regard to independent cinema, see Sue Clayton and 
Jonathan Curling, ‘On Authorship’, Screen Spring 1 979, vol 
20 no 1, pp 35-61. 

^ Walter Ong, Orality and Literacy: The Technologicing of the 
Word, London, Methuen New Accents, 1982, pp 79-83, 
traces a similar concern over writing back to Plato’s 

2) the right to record broadcast or cable programmes 
off air for educational purposes without cost, unless 
restricted by a certified licensing scheme. 

3) the right to record broadcast or cable programmes 
for designated non-profit archives. 

We hope to analyse the broader significance of this 
legislation in a future issue. 

The second part of this one is devoted to a debate 
between the co-authors of The Classical Hollywood 
Cinema, Kristin Thompson, Janet Staiger and 
David Bordwell, who is also author of Narration in 
the Fiction Film, with Barry King, a London-based 
member of the Screen Editorial Board who recently 
reviewed those titles along with Edward Branigan’s 
Point of View in the Cinema. 

In reconsidering the arguments of two extensive 
studies of the fiction film, these articles range widely 
over questions of style, technical innovation, mode 
of production, narration and human cognition. At 
stake are the boundaries of formal difference and the 
dynamics of their determination, as well as the very 

relevance of theories of ideology and the subject to 
such concerns. 

The replies raise a number of further questions, 
among them the unity of the authors’ work within 
a larger project of cinema poetics, and any corres- 
ponding (or indeed opposing) unity of approach 
within this journal. Similar questions are debated 
about the research facilities and funding available in 
the US and the UK, and their possible influence on 
national approaches to film scholarship. It is a vast 
and at times fiercely contested discussion, and we 
anticipate further contributions on the many ques- 
tions it opens. 

During the production of this issue we were very 
sorry to learn of the death of Claire Johnston, a 
founding figure of feminist film criticism and a 
valued member of the Society and the Screen 
Editorial Board over many years. We extend our 
sympathies to her family, and to her friends and 
colleagues in the field. 






* David B Hopkins, 
‘Ideas, Their Time Has 
Come: An Argument 
and a Proposal for 
Copyrighting Ideas’, 
Albany Lav! Review, 
vol 46 no 2, Winter, 
1982, p 453. 

THE RECENT EVOLUTION of copyright law in the 
United States is particularly rich in contradictions because of the pres- 
sure exerted by new information technologies and by the requirements 
of an information economy - that is, an economy based in the progres- 
sive commodification of information. In so far as this process is 
mediated through the law relating to intellectual property, it 
exacerbates the contradiction between the principles of limited 
monopoly rights and the public availability of ideas. 

The increasing extension of capitalist relations of production to the 
forms of intellectual work (both through the integration of workers into 
an industrial mode of intellectual production, and in the consequent 
breaking down of an unmediated relation between labour and product) 
has made ever less tenable those categories of copyright law, such as 
‘author’, ‘work’, ‘copy’, ‘expression’, etc, which have been derived from 
a romantic aesthetic of free creativity. Indeed, there is some pressure 
now from right-wing jurists for an even fuller integration of copyright 
law into the service of the market, through a dismantling of the funda- 
mental doctrinal distinction between ideas and expressions. Hopkins 
argues, for example, that the law at present provides no protection for 
such things as ‘a game concept, the system that goes into legal papers, or 
the ideas comprising the methodology or processes adopted within a 
computer program’ ‘, and, more generally, that ‘the very fact that ideas 
are free creates a disincentive to the development of ideas. It is only 
when people can fully exploit the benefits of their ideas and receive pro- 


tection in these endeavours that they will donate the product of their 
work process to the public domain.’^ 

American copyright law as it now stands protects (1) expressions of 
ideas in (2) works of authorship fixed in (3) copies or phonorecords. The 
‘copy’ thus stands at the end of a double set of substance/expression rela- 
tions, andu may in turn form the starting point for the production of a 
further chain of materialisations of the ‘work’. 

Title 17 of the US legal code defines ‘copies’ as ‘material objects, 
other than phonorecords, in which a work is fixed by any method now 
known or later developed, and from which the work can be perceived, 
reproduced, or otherwise communicated, either directly or with the aid 
of a machine or device. The term “copies” includes the material object, 
other than a phonorecord, in which the work is first fixed.’ ^ In the case 
of a painting, for example, this object is what we would usually think of 
as the ‘original’^ the ‘original’ - the painting, the manuscript, the master 
tape - is already a ‘copy’ of the (immaterial) ‘work’. The ‘work’ that is 
copied thus has no separate material form, and no existence prior to the 
moment of its fixation in the copy; but as an elaborated expression it is 
also different in kind from the ‘ideas’ it expresses. It has a peculiar mode 
of existence which is neither concrete nor fully abstract. 

What the disjunction between work and copy here expresses is essen- 
tially the existence of two distinct sets of property rights. Rights of 
ownership in the ‘material object’, which a painter sells to the purchaser 
of a painting, do not necessarily carry with them the rights to make and 
sell reproductions of the object. The law does not think the difference 
between work and copy as a temporal disjunction, because of the legal 
fiction that the ‘creation’ of the work takes place at the moment when it 
is first fixed in a copy dr phonorecord'’; but the disjunction does mean 
that the ‘fixing’ of the work need not be singular: ‘the same work may 
... be embodied in a range of “copies” including periodicals, computer 
punch cards, microfilm, tape recordings, and the like.’’ 

Neither work nor copy, however, can be defined in terms of material 
or structural self-identity. Rather, they are defined in relation to an 
intentional act, an act of human will. Thus proof of copying is not given 
by the simple identity of two works, since in addition a copyright holder 
must establish that a deliberate act of copying has taken place, or at least 
establish its physical possibility and the likelihood of its occurrence. 
Conversely, if it can be demonstrated that an independent act of creation 
has taken place, then two identical works may each be entitled to copy- 
right protection. Hence Judge Learned Hand’s famous dictum in Shel- 
don v Metro-Goldwyn Pictures Corp (1936): 

^ ibid, p 446. 

’ United Slates Code 
Service, vol 1 7, 
paragraph 101. 

■* ibid. 

’ Neil Boorstyn, 
Copyright Law, 
Rochester, The 
Lawyers Co-operative 
Publishing Company 
and Barncroft-Whitney 
Co, 1981, paragraph 
' 2.4. 

Borrowed the work must indeed not be, for a plagiarist is not himself pro 
tanto an ^author'; but if by so?ne magic a man who had never known it were 
to compose anew Keats’s Ode on a Grecian Urn, he would be an ’author’, 
and, if he copyrighted it, others may not copy that poem, though they might 
of course copy Keats’s. 

® American Jurisprudence 
vol 1 8, 2d: Copyright 
and Literary Property, 
paragraph 41. 

Although an ‘author’ may not be a ‘plagiarist’, s/he is not defined as a 
pure originator. The ‘productivity’ component of fair use doctrine 
clearly recognises that textual production is a cumulative process, neces- 
sarily involving a relation to the body of preceding texts; and certain 
classes of work - such as a ‘translation, musical arrangement, dramatiza- 
tion, fictionalization, motion picture version, sound recording, art 
reproduction, abridgement, condensation’, or ‘a work consisting of 
editorial revisions, annotations, elaborations, or other modifications’ 
may be classified both as a ‘derivative work’ and as an ‘original work of 
authorship’ with entitlement to copyright protection.* 

These distinctions are already finely drawn; but the development in 
the twentieth century of mechanical and electronic technologies which 
multiply the number and form of the mediations involved in reproduc- 
tion has made the distinction between work and copy even more prob- 
lematical. One key case, which has had continuing consequences for the 
status of computer software, is the 1908 White-Smith Music Publishing 
Co V Apollo Co, where the court refused to recognise player pianos and 
the music rolls used to reproduce musical compositions as producing a 
‘copy’ of those compositions. At issue was partly the question of 
whether perforations in a music roll could be considered an ‘intelligible 
notation’, but also the very possibility of distinguishing between a tune 
(an organisation of sounds) and its physical embodiment in sounds (the 
‘copy’). In rejecting this possibility theVourt found that the organisation 
of sounds produced by the player piano was the tune itself, the ‘original 
work of authorship’, rather than a secondary copy or embodiment of it. 

Part of the difficulty with the distinction between work and copy is 
that the categorieis are purely positional. Thus the performance of a 
work can in its turn become eligible for copyright - that is, it can itself 
be considered an original work. One area of particular complexity in this 
respect is broadcast radio and television, where multiple realisations of a 
work may occur - in the studio, in transmission, in retransmission, and 
in the home or other place of reception. It is the retransmission process 
that has raised the most difficult, because commercially most signifi- 
cant, problems. Cable systems, like VCRs, have the ambivalent status of 
being both a reception and a retransmission apparatus, and while the 
1968 Supreme Court decision on cable television upheld the operators’ 
claim merely to be enhancing a signal, the imposition of royalties in the 
1976 Copyright Act implicitly recognises that an act of reproduction 
does take place in cable retransmission. 

If the fixing of the ‘work’ in the ‘copy’ is capable of taking a number of 
materially distinct forms, then it may be the case that these forms do not 
physically resemble each other or the ‘work’. Copying, that is, need not 
imply iconic likeness. Thus an oral text may be copied in written form; 
and written texts may be reproduced as a set of electronic impulses, or in 
the medium of film. It is at this point that the distinction between work 
and copy shades offinto that between ideas and their expression. This is 
clearly evident in the case of those practical schemata that we might call 

‘scores’, where an act of realisation can be thought either as a copy of a 
work or as a simple use of unprotected ideas. The following of a recipe in 
cooking a meal, for instance, could be thought of as a performance equi- 
valent in kind to the following of a musical score, but only the latter is 
protected as a ‘work’; the recipe is unprotected because it is categorised 
as an ‘idea’ (a ‘method’ or an ‘art’). Similarly, architectural blueprints 
have an ambivalent status as either ideas or the expression of ideas. In 
the USA, the construction of a building in accordance with a design is 
understood as an expression of the ideas contained in the design; there 
is, therefore, little protection for anything but the material form of the 
design, except in the case of decorative additions.’ In the UK, by con- 
trast, the building may be thought of as a copy of the blueprint, and this 
means that the design is given much greater solidity as an ‘expression’ 
(and so as a ‘work’- that can be protected).® 

The basic purpose of copyright law is at once to restrict the com- 
pletely free circulation of intellectual products, and to ensure the free 
accessibility of any ‘idea, procedure, process, system, method of opera- 
tion, concept, principle, or discovery’ - that is, of any ‘intellectual con- 
ception apart from the thing produced’’. The distinction is clearly 
problematical, however, both in its attempt to protect expressions of 
ideas without implicating the ideas themselves, and in its implicit attri- 
bution of an ontological status to the two terms. Judge Learned Hand 
put concisely the dilemma of the necessity apd the impossibility of the 
distinction in his formulation of an abstraction test: 

’ Neil Boorstyn, op cit, 
paragraph 2.25. 

® LB Plastics v Szcish 
Products (\979)j cited 
in Christopher} 
Millard, Legal 
Protection of Computer 
Programs and Data^ 
London, Sweet and 
Maxwell, 1985, p 18. 

^ United Slates Code 
ServicCy vol 17, 
paragraph 101. 

Nichols V Universal 
Pictures CoTp (1930). 

upon any work, and especially upon a play, a great number of patterns of 
increasing generality will fit equally well as more and more of the incident is 
left out. The last may perhaps be no more than the most general statement of 
what the play is about, and at times might consist only of its title; but there is 
a point in this series of abstractions where they are no longer protected, since 
otherwise the playwright could prevent the use of his ‘ideas', to which, apart 
from their expression, his property is never extended. Nobody has ever been 
able to fix that boundary, and nobody ever caw.'® 

The corollary that Learned Hand doesn’t draw here is that every level of 
abstraction functions as the ‘idea’ in relation to all less abstract levels; 
and that any level can thus function either as that of ‘ideas’ or of ‘expres- 
sion’, depending on its relation to other levels of abstraction. 

This fundamental aporia gives rise to recurrent ambiguities. Musical 
works, for example, can exist in the form either of written notation or of 
a ‘phonorecord’, that is, the fixation of a performance. The latter can 
represent two distinct types of copyrightable work: a musical composi- 
tion, or a particular embodiment of the musical composition in sound. 
The performance can thus either be equivalent to a notation, or it can be 
the realisation of a notation. In the case of the musical composition, 
what is copyrightable is an organisation of sounds, not a particular 
realisation of them. ‘Sounds’ are thus understood as immaterial. 

’ * Neil Boorstyn, op cit, 
paragraph 2.9. 

Sheldon V Metro- 
Goldhiyn Pictures, Inc 
(1936); Universal 
Pictures Co v Harold 
Lloyd Corp (1947); 
Hoehling v Universal 
City Studios, Inc 

* ^ Wa rner B ros Pictu res, 
Inc V Columbia 
Broadcasting System, 

■ 7nc(1954). 

*'* Neil Boorstyn, op cit, 
paragraph 2.22, citing 
Herbert Rosenthal 
Jetvelry Corp v 
Baker v Selden (1879). 

* ^ Synercom Technology, 
Inc V University 
Computing Co (1979). 

abstract, and it is arguable that, in so far as the terms retain any mean- 
ing, copyright here resides in an ‘idea’ rather than an ‘expression’. In the 
case of the performance, by contrast, it is a particular material realisa- 
tion of these sounds (but not the ‘tangible medium of expression’) which 
is subject to copyright - and there can thus be multiple copyrights in 
authorised performances of a single work. The concepts of ‘sound’ and 
‘performance’ can therefore represent both a content plane and an 
expression plane. A similar ambiguity holds in relation to pantomimes 
and choreographic works, which ‘can be fixed in any tangible medium 
of expression such as film, video tape, dance notation (Laban system), 
diagram, or verbal description’", and where, as well as being equivalent 
forms of notation of an ‘idea’, any one of these could act as the ‘expres- 
sion’ or realisation of any other. 

Literary works are somewhat more amenable to the distinction 
between ideas and expressions than are nonverbal scores, since the oper- 
ations of paraphrase and summary which underlie it originate in literary 
and philosophical pedagogies. Here the relevant distinctions are 
between an outline or theme or locale or generic plot structure (the scene 
a faire), on the one hand, and a ‘distinctive treatment’ of these on the 
other.'^ Characters may be afforded protection if they exist in visual or 
graphic form and can be held to embody both physical and conceptual 
characteristics, but literary characters are usually held to be an abstract- 
able element of consent. The test used in the 1954 ‘Sam Spade’ case was 
that ‘unless the character really constitutes the story being told’ and is 
not ‘only the chessman in the game of telling the story he is not within 
the area of protection afforded by the copyright’*^. In theoretical terms 
the distinctions being made here are untenable; but what they seem to 
reflect is an identification of ‘ideas’ with verbal structure and of ‘expres- 
sion’ with the iconic or the visual. 

Copyright doctrine does at times recognise that the distinction 
between ideas and expression cannot be universally drawn. In cases 
involving such things as blank accounting forms which are an integral 
part of a book-keeping system, the principle is that where an ‘art’ (a sys- 
tem, an idea, a method) cannot be used without copying, then copying 
for such ‘use’ is no infringement (although copying an expression for the 
purpose of ‘explanation’ would be); ‘where the idea and its expression 
are indistinguishable, this inseparability will permit copying of the 
expression. Otherwise “protecting the expression in such circumstances 
would confer a monopoly of the idea . . . free of the conditions and limit- 
ations imposed by the patent law.” This should mean, however, that 
no copyright protection can be afforded where ideas and expression 
fully merge. Certainly there has been a ruling to this effect in the case of 
computer software*^; and it is arguable that it would apply to all iconic 
expressions, and certainly to those which are non-figurative (this is per- 
haps the reason why such expressions can receive additional protection 
as ‘works of art’). 

The case of cartography can usefully help extend the analysis of these 

categories. Here the problem is that the more accurate maps become, the 
more they will resemble each other, and it would seem, on the face of it, 
difficulty to grant copyright to two more or less identical maps. A num- 
ber of court decisions since the last war therefore began to introduce the 
criterion of novelty in order to deny copyright to maps. 

Copyright doctrine requires, however, only that works of authorship 
be original; and this is explicitly distinguished from novelty.” The cri- 
terion of originality demands no more than that the work display ‘some- 
thing irreducible, something which is one man’s alone’*®. This means 
that what is protected is not distinctiveness of expression - not an inher- 
ent difference from other expressions - but only (as with the text that 
re-produces, without copying, the ‘Ode on a Grecian Urn’) whatever 
distinctiveness derives from the originality of an act of authoring. 

Authorship as origin is, then, the most fundamental category of copy- 
right law, in relation to which all other categories are secondary. It is the 
principle that founds both the work and the copy in their respective acts; 
both the idea and its expression. But this principle in turn requires fur- 
ther analysis. 

The concepts of work and author are in a tautological relation to each 
other. A work of authorship is anything which is the product of an 
author, whereas ‘ “author”, in a constitutional sense, means “he to 
whom anything owes its origin; originator; maker” ’*®. This tautology is 
not purely uninformative, however, because it means that a work can 
always be traced back to an originary principle, and that it is this prin- 
ciple, which defines the specificity of the work. Thus various judicial 
decisions have held ‘that authorship implies that there has been put into 
a production something meritorious from' the author’s own mind; that 
the product embodies thought of the author, as well as the thought of 
others; and that it would not have found existence in the form presented, 
but for the distinctive individuality of mind from which it sprang’^®. 

Authorship is thus a general principle of differentiation; but because 
this principle can only readily be grasped in terms of differentiations 
between works, the concept of originality has been defined in two 
slightly different ways. In the first, the ‘something irreducible’ that 
marks originality is located in the work itself; the law is interpreted as 
making a minimal demand for some more than ‘merely trivial’ variation 
to distinguish the work from other works.^* Because this is a minimal 
requirement, however, ‘copyrighted matter need not be strikingly 
unique or novel, and any distinguishable variation resulting from an 
author’s independent creative effort will sufiice.’^^ Thus slight changes 
in appearance from the products of competitors have been held suffi- 
cient to attract copyright protection to dolls or stuffed animals, or to a 
sketch based on Paddington Bear. Moreover, the ‘distinguishable varia- 
tion’ need not even be perceptible to an untrained observer. The Court 
in Gracen v Bradford Exchange (1983) commented that ‘artistic 
originality may inhere in a detail, a nuance, a shading too small to be 
apprehended by a judge’; and further that 




Cartography, and 
Copyright’, Wra York 
University Lata 
Revievisto 38, 1963, 
pp 283fr. 

*^ Lin-Brook Builders 
Hardtaare v Gertler 

*® Bleistein V Donaldson 
Lithographing Co 

*® Burrovi-Giles 
Lithographic Co v 
Goldstein v California 

National Telegraph 
Nevts Co V Western 
Union Telegraph Co 

^* Kamar International, 
Inc V Russ Bessie and 
Co (1981). 


Jurisprudence \o\ 18, 
2d, paragraph 19. 


JosU Gracen v The 
Bradford Exchange 
(1983). The judge in 
the case was the ‘law 
and economics’ 
theorist Richard 

RF Whale and Jeremy 
J Phillips, Whale On 
Copyright, 1971, rpt 
O.xford, ESC 
Publishing Limited, 
1983, p 39. 

ibid, p 21. 

Nimmer on Copyright, 
cited in Neil 
Boorstyn, op cit, 
paragraph 2.12. 

since a contemporary school of art known as ‘Super Realism' attempts to 
make paintings that are indistinguishable to the eye from colour photo- 
graphs, and these paintings command high prices, buyers niust find some- 
thing ‘original' in them. The court further noted that since much Renaissance 
painting is meticulously representational, it is therefore in a sense (but not an 
aesthetic sense) less ‘original' than Cubistn or Abstract Expressionism.^^ 

But of course the concept of originality does not depend on the relation 
between signifier and signified; and this reading seems to me to demon- 
strate the limits of this first line of argument. 

The second way of defining originality is without reference to the 
status of the product. Whale puts it this way: ‘Originality is not to be 
equated with the creation of something which had not hitherto existed; 
it is the word used to describe the causal relationship between an author 
and the material form in which a work is embodied. This definition of 
originality as a causal relationship has the advantage of being able to 
explain certain apparent anomalies in copyright law, above all the 
importance given to the moment and the process of fixation - of material 
realisation - rather than the moment and process of creation. (Thus, for 
example, ‘the “author” of a photograph is the owner of the film upon 
which it is taken; and if a musician composes an impromptu tune which 
another records, it is the person upon whose tape the recording is made 
who becomes first owner of copyright in the sound recording.’”) It helps 
explain, too, the distinction made in relation to the separate copyright 
category of the ‘work of art’, which requires both originality and ‘some 
creative authorship in its delineation or form’, where, as Nimmer puts 
it, ‘creativity refers to the nature of the work itself, originality refers to 
the nature of the author’s contribution to the work’.^® 

This conception of originality is at once broader, in that it refers to 
persons other than the ‘creator’ of the work, and narrower, in that it 
excludes the nature of the work itself from consideration except in so far 
as this is the effect of its origin. Again, however, the concept can be 
understood in two rather different ways, reflecting a fundamental ambi- 
guity in the social function of copyright law. These are, briefly, in terms 
of an investment of capital, or of an investment of labour. 

On the one hand, ever since its origins in the 1709 Statute of Queen 
Anne, which protected the commercial exploitation of printed books, 
copyright law has vested the right of reproduction in copyright owners 
rather than directly in authors. With the industrialisation of the 
production of information the non-coincidence between the two has 
become commercially crucial. Cornish notes that the British Copyright 
Act of 1911 

gave the producers of sound recordings their own exclusive right to prevent 
reproductions of their recordings (and, as the courts later held, also to prevent 
public performances of them). The right was indiscriminately labelled copy- 
right, even though it was conferred, not upon the executant artist whose per- 


formance was recorded, but upon the business which organized the 

And Brecht has written extensively on the subordination of the ‘author’ 
and the author’s property rights to the economic requirements of the 
film industry, in such a way that the representative of capital, the pro- 
ducer, effectively takes over the author function for legal purposes.^® 

On the other hand, copyright doctrine has often taken the approach of 
directly protecting an investment of labour rather than the work which 
is the ostensibly protected object. This has been the case, for example, 
with works such as computer databases and compilations where there is 
no ‘organisation of ideas’ to protect. Thus the Court in National Busi- 
ness Lists V Dun and Bradstreet, Inc (1982) decided that 

compilations such as Dun and Bradstreet’s have value because the compiler 
has collected data which otherwise would not be available. The compiler's 
contribution to knowledge normally is the collection of the information, not 
its arrangement. If his protection is limited solely to the form of expression, 
the economic incentives underlying the copyright law are largely swept away. 

Similarly, in the case of cartography the possibility of protection of vir- 
tually identical maps is seen to reside in the labour of production which has 
gone into them. There is indeed debate about whether consulting and 
conflating a number of previous maps constitutes an appropriate kind and 
degree of work, or whether a cartographer must actually undertake 
research in the field; but in either case the category of originality is reduced 
to labour. As Whicher writes: ‘when the creative process is re-examined by 
the wisdom of judicial hindsight, it is, like a conjuror’s trick that has been 
explained to the children, almost always a disappointment. There is, we 
discover, no magic to it after all. It’s only work.’^’ 

What this might mean is that the concept of originality, this most funda- 
mental category of the copyright system, can be rethought in materialist 
terms . The ‘causal relationship between an author and the material form in 
which a work is embodied’ would be an investment of labour power. 
Understood as work, the concept of authorship could then be freed of its 
exclusively individualistic connotations: the author function (which 
would be a moment in a system of production) could equally be performed 
by a team, by a production crew, by a group of collaborators; and the 
ideology of free creativity could be displaced, but in terms that are derived 
internally (if critically) from the existing structure of copyright law. 

WR Cornish, 
Intellectual Property: 
Patents, Coypright, 
Trade Marks and 
Allied Rights, London, 
Sweet and Maxwell, 
1981, pp 299-300. 

Bertold Brecht, ‘Der 
Schrijien zur Literatur 
und Kunst /, 
Gesammelte Werke, 

18, Frankfurt, 
Suhrkamp, 1957. 
There is a paraphrase 
and commentary on 
this essay in John 
Frow, ‘Film, 
Production and the 
Law: Brecht’s 
Australian Journal of 
Cultural Studies, vol 2 
no l,May, 1984, pp 
3 - 22 . 

John F Whicher, op 
cit, p 295. 


That a set of economic categories, and the ideological contradictions 
attendant upon them, can so readily be derived from the structure of 
copyright doctrine bears witness to its practical closeness to economic 


Untied Stares Code 
Service, vol 17, 
paragraph 102(a). 

ibid, paragraph 101. 


processes and economic contradictions. The specificity (the ‘autonomy’) 
of copyright doctrine is not opposed to, but is precisely the effect of, its 
continuous mediation of antagonistic economic and social interests. 

The most acute provocation to the coherence of copyright theory has 
come in the last two decades from the development of electronic infor- 
mation storage and retrieval systems which, in vastly multiplying the 
possibilities of deployment of complex bodies of information, have simi- 
larly expanded the industrial importance of the control of information 
and have increased the stakes in its ownership and protection. 

There are significant difficulties, however, in applying traditional 
conceptions of intellectual property to electronic systems. Most infor- 
mation stored electronically may, for example, never be expressed in a 
permanent fixed form but will be used and stored in a form which, like a 
television display, is transient and would therefore not attract copyright 
protection. Information stored and modified within an electronic net- 
work with multiple access may be shaped and reshaped in such a way 
that it is difficult to determine an ‘author’. Programs may exist in a form 
which is not intelligible to a human user, but is designed only for inter- 
action with a machine (and machines may in turn have the capacity to 
modify the program). And information in electronic form may be 
‘stolen’ by means of a transfer which involves no alienation of property; 
proof of theft may therefore be extremely difficult, or it may be trivial, if 
what is stolen is deemed to be, say, a print-out rather than the informa- 
tion it contains - which may in any case have no easily assignable value. 
Moreover, there is a problem with the apparent solution to this, which is 
(as Posner and other ‘law and economics’ theorists do) to understand 
intellectual property and its protection by direct analogy with tangible 
property. Any such solution plays down the contradiction between 
rights of exclusive use and the social interest in keeping information in 
the public domain. 

Some of these difficulties are resolved by the 1976 Copyright Act’s 
careful wording of the fixation requirement to protect ‘original works of 
authorship fixed in any tangible medium of expression, now known or 
later developed, from which they can be perceived, reproduced, or 
otherwise communicated, either directly or with the aid of a machine or 
device’^®, and by the amendment of the Act’s definitional section in 
1980 by the Computer Software Protection Act to include a definition of 
a computer program as ‘a set of statements or instructions to be used 
directly or indirectly in a computer in order to bring about a certain 
result’^'. In addition, the House Report on this legislation made it clear 
that, as provided for in Section 102(b), which bars copyright protection 
for ideas, procedures and methods of operation, ‘copyright in a 
computer program does not extend protection to the methodology or 
processes adopted by the programmer. Only the expression adopted by 
the programmer is protectible, and not the methods or processes 
embodied in the program.’’^ 


Nevertheless, these stipulations still leave a number of obstacles to 
interpretation. These concern, in particular, the question of whether 
protection should be available for programs in object code (that is, pro- 
grams written in a form which translates a programming language, or 
sourqe code, into a machine-readable form), and for the operating pro- 
grams which control the hardware functions, and which would, there- 
fore, seem to be equivalent to machine parts and so eligible for patent 
rather than copyright protection; but which are in practice often hard to 
distinguish from the application programs that produce copyrightable 

There are two contradictory arguments, which don’t quite meet head- 
on, in play here. On the one hand, programs in object code and applica- 
tion programs are thought to be fully protected by the Act’s use of the 
word ‘directly’ to describe the use of instructions in a computer. On the 
other hand, it is argued that it is not possible to differentiate ‘ideas’ from 
‘expressions’ at the level of the object code, and that either the present 
Act, or present interpretations of the Act, have gone too far in restricting 
the availability of information. A couple of key cases may serve to illus- 
trate what is at issue here. 

Sytiercom Technology, Inc v University Computing Co (1978, 1979) 
involved a suit for copyright infringement of instruction manuals and 
input formats used with a computer program designed to solve certain 
engineering problems. The central question in the case, relating only to 
the input formats, was ‘whether [the defendant] plagiarized Synercom’s 
idea or its expression. If the idea is the sequence and ordering of data, 
there was no infringement. If sequencing and ordering of data was, 
however, expression, it follows that [defendants’s] preprocessor pro- 
gram infringed ’ This raises the further problem that ‘if sequencing 

and ordering is expression, what separable idea is expressed?’ The court 

’Idea’ or ‘expression’? 
Diagram of the 
translation process in 

cf Richard H Stern, 
‘Another Look at 
Copyright Protection 
of Software: Did the 
1980 Act do Anything 
for Object Code?’, 
Journal, vol 3 no 1, 
Fall, 1981, pp 16-17: 
‘To be protected 
under copyright law, 
an alleged copy must 
be intended to 
communicate some 
message intelligible to 
human beings, even if 
they need a machine 
to aid in the 
communication. That 
an object intended for 
a utilitarian purpose 
may be made to 
disgorge something 
intelligible to human 
beings if placed in a 
proper machine is 
probably insufficient. 
By these criteria, 
object code is not a 
/ copy of a work of 
authorship, and is 
thus not directly 
protected by 
copyright, nor 
indirectly protected 
under the theory that 
it is an infringement 
of a copyright based in 
the source program.’ 

finally held that the input format was in fact the expressed idea, which 
could, therefore, be freely copied. 

Data Cash Systems, Inc vjfS and A Group, Inc (1979, 1980) followed 
this ruling to hold that an object code could not be copyrighted. The 
case involved a computer program that instructed a computer how to 
play chess. The instructions were translated into programming 
language (the source^program) and then into machine language (the 
assembly program). The assembly program in turn was used to generate 
the object program, or Read Only Memory (ROM), which directly 
commands a series of electrical impulses and which is integrated with 
the computer’s circuitry. The Court held that the ROM was ‘a 
mechanical tool, or a machine part, or the mechanical embodiment of 
the source program, but not a copy of it’, and so not a ‘writing’. The 
copyright protection afforded by the 1976 Copyright Act was therefore 
held to apply to computer programs in their flow chart, source, and 
assembly phases, but not in their ROM or object phase.” 

What is in question in these and similar cases is the process of trans- 
lation between different levels of the computational process, and hence 
the more general problem of distinguishing ideas and expressions. Let 
me define these translation processes schematically: 

(1) from one high-level language to another. Here, the judge in Synercom 
ruled that ‘it is a's clear an infringement to translate a computer program 
from, for example, FORTRAN to ALGOL, as it is to translate a novel 
or play from English to French. In each case the substance of the expres- 
sion (if one may speak in such contradictory language) is the same 
between original and copy, with only the external manifestation of the 
expression changing.’ 

(2) from flow chart to source code. The same judge felt that this would 
equally constitute an infringement of copyright; but he drew the line at 
granting protection for 

(3) translation from a general statement of the program to a source code: 

Here the similarity to literary translation ends. The preparation of a compu- 
ter program in any language from a general description of the problem to be 
solved. . . is very dissimilar to the translation of a literary work, or to the 
translation of a program from one language to another. In most cases, the 
formulation of the problem in sufficient detail and with sufficient precision 
to enable it to be converted into an unambiguous set of computer instructions 
requires substantial imagination, creativity, independent thought, and exer- 
cise of discretion, and the resulting program can in no way be said to be 
merely a copy or version of the program statement. The program and the 
statement are so different, both in physical characteristics and in intended 
purpose, that they are really two different expressions of the same idea, 
rather than two different versions of the same expression. 

(4) from source code to object code. Perhaps the most forceful argu- 
ments here are those made in 1979 by Commissioner Hersey in his 

dissenting report to the National Commission on New Technological 
Uses of Copyrighted Works (CONTU). Kersey’s view is that, unlike 
sets of instructions, ‘in the case of computer programs the instructiom 
themselves eventually become an essential part of the machinery that 
produces the results'^* and ‘a program, once it enters a computer and is 
activated, does not communicate information of its own, intelligible to a 
human being. It utters work.’’^ The proper analogy for a computer pro- 
gram is not with a film or a phonorecord, but with such things as 
magnetised bank cards, or a cam (a gearing device) that instructs other 
machinery. Above all. Kersey stresses that programs are neither ‘works’ 
nor ‘copies’: ‘a program, when keyed or run into a computer, is trans- 
formed by a compiler program into a purely machine state. The term 
copy is meaningless for the reason that in this transformation the means 
of expression of the original work become totally irrelevant. All that 
matters is the program’s functional useJ^* If this is arguably true of a 
source program or code, it is a fortiori arguably true of an object code. 
Thus the judge in Data Cash argued that, ‘by analogy with a building 
constructed from architectural plans, object code constituted the 
physical embodiment of a computer program. Just as, under American 
law, a building is not a “copy” of its plans, so too a ROM is not a copy of 
a source program’, and so does not qualify for the status of a ‘writing’.” 
The relations at work in these four instances are those between expres- 
sions; between ideas and expressions; and between abstract and concrete 
forms'^if an expression (with the complication of a Kjelmslevian distinc- 
tion, in (1), between the form and the substance of an expression). It is 
clear that it is the fourth class, the relations between source code and 
object code, that resists explanation in terms of the traditional cate- 
gories; and subsequent rulings have tended either to extend a blanket 
protection to object code, or to draw the line between idea and expres- 
sion pragmatically, taking into account the balance between competi- 
tion and protection. Apple Computer, Inc v Franklin Computer Corp 
(1980, 1982) is a particularly interesting example of the interpretive 
dilemma posed to the courts, since it produced diametrically opposed 
readings in the district court hearing and on appeal. The district court 
found that ‘it is not clear whether the program-designer’s idea of the 
operating system program, the source program, or the ROM is the 
“original work of authorship.” It is not surprising that this should be 
hard to determine, because at each stage major transformations in the 
structure of the “program” take place.’ The judge decided that, rather 
than understanding object code as a ‘language of description’, ‘it may be 
more accurate to say that operating systems are an essential element of 
the machine, if not an essential part of the machine that makes it work. 
Similarly, it may be more accurate to say that object code in its binary 
form or chip form is a useful version of the machine’s electrical pulse.’ 
The Court of Appeals reversed this decision, however, to hold that 
object code could be afforded protection as a ‘literary work’: ‘The court 
rejected the defendant’s argument that the ROM-embedded object code 


^ * Final Report of ike 
National Commission 
on Nezo Technological 
Uses of Copyrighted 
Works, Washington, 

Library of Congress, 

1979, p 28. (Italics as 
in original.) 

ibid, p 29. 

ibid, p 32. 

” Christopher J 
Millard, p 38. 


Anderson L Baldy III, 
‘Computer Copyright 
Law: An Emerging 
Form of Protection 
for Object Code 
Software After 
“Apple V Franklin” 
Journal, vol 5 no 2, 
Fall, 1984,p248. 

\VR Cornish, op cit, p 

Apple Compuler, Inc v 
Computer Edge Ply 
Ltd and Suss (1984), 
cited in Christopher J 
Millard, op cit, p 71. 

cf Christopher J 
Millard, op cit, pp 

Glenn J MacGrady, 
‘Protection of 
Computer Software - 
I An Update and 
Practical Synthesis, 
Houston Lata Revieta, 
vol 20 no 4, July, 
1983, p 1046. 

Final Report of the 
National Commission 
on Neta Technological 
Uses of Copyrighted 
Works, p 26. 

ibid, p 34. 

was an uncopyrightable idea, noting that since other object code pro- 
grams could be written to perform the same functions, Apple’s object 
code was merely an expression of the idea. . . . 

Cornish has suggested that ‘the computer’s work of “translating” the 
program from a high-level to a machine language is a separate copyright 
activity’.” An Australian court, dealing with a parallel case, decided 
that object codes are protected as ‘adaptations of the protected source 
codes’, that is, as ‘translations’.'*® And some recent findings in relation 
to video games have concluded that ROMs constitute a ‘tangible 
medium’ for the fixing of expressions of ideas.'*' This plethora of differ- 
ent solutions perhaps indicates that there is no good solution to the 
problem. That it is not a marginal problem but absolutely central to the 
question of protection of computer software is indicated by the argu- 
ment made in Williams v Artie (1982) that to exclude object code from 
protection ‘would afford an unlimited loophole by which infringement 
of a computer program is limited to copying of the computer program 
text but not to duplication of a computer program fixed on a silicon 
chip’. In practical terms this means that the protection extended to one 
level of coding can be thwarted simply by copying directly from the 
level of the object code. 

It is in line with this that MacGrady argues for the inadequacy of 
existing copyright protection for computer programs, pointing out that 
‘there is a crucial distinction between the design or algorithm of a pro- 
gram and the coding of a program. Copyright law will protect only the 
particular code expression chosen by the coding programmer or a cod- 
ing expression substantially similar thereto; it will not protect the design 
itself.’*^ Others, however, have warned against a tendency on the part of 
both legislators and courts to afford precisely such a protection to design 
or ‘ideas’. In his concurring opinion for the CONTU report Nimmer 
wrote: ‘What is most troubling about the Commission’s recommenda- 
tion of open-ended copyright protection for all computer software is its 
failure to articulate any rationale which would not equally justify copy- 
right protection for the tangible expression of any and all original 
ideas’*'; and Hersey bluntly stated that the Commission ‘may well have 
opened the way for covert protection, in the name of copyright, of the 
underlying mechanical idea or ideas of a program, rather than of its ori- 
ginal means of expression’**. 

There is another dimension to the question of object code, however, 
which may be equally significant; it concerns its status as a non-human 
communication. Since computer programs are usually written in a 
source code and then compiled by means of a separate program into 
object code, there is a question as to whether the program is in fact 
authored by a person rather than by a machine. Computers have been 
used extensively in graphic design, in musical composition, and in the 
production of written texts. In the case of minimal human intervention, 
who should be considered their ‘author’ for copyright purposes? More 
importantly, by what criteria do we distinguish between human and 

‘The PUZ’; program for an Atari ST computer game. 

artificial intelligence for copyright purposes? 

Millard points out that, not only can computers produce output 
which is unpredictable and which can thus to some extent ‘emulate the 

Christopher J 
Millard, p 26. 

ibid^ pp 26-27. 

ibid, pp 27-28. 

ibid, p 28. 

ibid, pp 28-29. 

vagaries and spontaneity of the human brain’'*’, they can also simulate 
‘free will’ in their ability to avoid predictability and determinism - for 
example, by selecting random numbers in an infinite progression, and 
using these (for example) for the production of musical texts. Millard 
follows Butler in suggesting four possible responses to this question. 
They are: 

(1) disallow copyright completely; (2) give authorship and copyright to the 
computer and its software or find authorship ‘shared’ between the 
AfrtificialJ IfntelligenceJ software and a human; (3) settle copyright upon the 
owner of the underlying AI software or the machine owner; or (4) create a fic- 
tional human author and assign its copyright to the AI software owner, the 
problem-specifier or the computer owtter either individually, jointly or in 

To these, Millard adds the option favoured by the 1976 UK Whitford 
Committee, of granting copyright to the person who provides the data, 
alone or jointly with the owner of the AI software; and the option 
favoured by the 1981 Green Paper Reform of the Law Relating to 
Copyright, Designs and Performers’ Protection, of granting copyright to 
the person responsible for running the data through the programmed 

All of these options are problematical. The first leaves the work 
unprotected. The second (machine or shared human and machine 
ownership) ‘would necessitate absurd legal gymnastics to accommodate 
established copyright principles. For example, how long should a com- 
puter be deemed to live for the purpose of fixing the term of protection 
for a work? Moreover, how could a computer assign or otherwise admin- 
ister its rights and how should it be rewarded for its “creative 
effort?” The third option of awarding copyright to the owner of the 
software or hardware ‘might be effectively to grant a monopoly over a 
process for producing a vast number of different works’, but also, like 
the last two options, it might fail to establish a ‘creative link’.'*® 

It is the legal fiction of the fourth option, then, that Millard concludes 
by finding most attractive. This option, of ‘creating a fictional author 
and assigning his or her copyright to other parties’, has the advantage 
that it ‘both ensures protection without violating the concept of creativ- 
ity as a distinctively human endeavour, and awards copyright protection 
to human beneficiaries’^’. 

Clearly it takes some rather hard work to preserve the conceptual 
priority of human creativity here. In fact, unlike most European law, 
where authors must be natural persons, Anglo-American law allows 
legal entities (corporations or partnerships, for example) to hold copy- 
right as an author. The legal and commercial conditions for dropping 
the insistence on individual human creativity as the source of copyright 
are fully in place, and the fiction is fully recognisable as such. 

What the fiction screens is the fact that copyright is part of a system of 

commodity production in which reproduction rights play a major part; 
and that computer software is a significant component of this system. 
One commentator put the number of programs written daily in the 
United States at 15,000^®, and the International Data Corporation esti- 
mated that in 1980 there were more than 4,300 companies in the soft- 
ware industry, with revenues of 13.14 billion dollars rising to 33.8 
billion dollars in 1984.^* This market is increasingly coming to be 
dominated by large corporations. Hence Commissioner Hersey’s 
expression of concern at the pressures for an extension of the copyright: 
‘Is it not evident’, he wrote, ‘that the big companies want, by availing 
themselves of every possible form of protection, to lock their software 
into their own hardware, while the independents want to be able to sell 
their programs for use in all the major lines of hardware?’^^ And in fact 
there have been significant extensions of the copyright and kindred 
areas of law as a result of such pressures in recent years, not only to 
protect computer software but to protect such things as the mask-work 
used in the production of silicon chips, or new plant varieties. At the 
same time, the possibilities of computer crime - embezzlement, piracy 
of software and data, theft of computer time, and sabotage - have 
increased exponentially, and in ways that existing intellectual property 
law is not fully adequate to prevent. De Sola Pool and Solomon identify 
the dilemma as this: ‘what protects the author or publisher is physical 
control of the text, for there is no count of its reproduction once it is out 
of his hands.’’’ The problem of control is exacerbated in the 
international marketplace, and it is with the question of transborder data 
flows that some of the most severe contradictions in the 
commodification of information come to light. Paradoxically, these 
involve a commitment to ‘free trade’ on the part of the countries and the 
transnational corporations with dominance in the international 
information industry, and a commitment to protectionist policies 
(privacy regulation, import controls, customs tariffs on data) on the part 
of Third World countries.’’ The struggle is one for control of a scarce 
commodity; but the concept of scarcity is inherently contradictory in 
this context. 

The contradiction arises from two features of an economy which is 
increasingly based on the commodification of information. The first is 
the augmented importance of information as a resource input to produc- 
tion - that is, the increase in the share of cultural and especially scientific 
capital in the organic composition of capital. Information technology, 
says Locksley, ‘enables 'a greater commercialization of culture where 
culture is not only a commodity but reinforces the conditions of general 
commodity production’”, and it ‘is the instrument that allows the 
resource and commodity of information to be developed and gain ascen- 
dancy. The current phase of capitalist development is one characterized 
by the elevation of information and its associated technology into the 
first division of key resources and commodities. Information is a new 
form of capital.’” As such, it is characterised by a transition from open. 

— 19 

” Walter E Schmidt, 

‘Legal Proprietary 
Interests in Computer 
Programs: The 

JurimetricSj vol 21 no 
4, Summer, 1981, pp 

” Glenn J MacGrady, 
op cit, p 1033. 

Final Report of the 
National Commission, 

” I de Sola Pool and R 
Solomon, ‘Intellectual 
Property and 
Transborder Data 
Flows’, Stanford 
Journal of 

International LaaXb, 
1980, p 121. 

” Sol Glasner, 
Corporations and 

Sovereignty’, in Anne 
W Branscomb (ed), 
Tozoard a Lazo of 

Netzoorks, The 
Science and 
Technology Section 
of the American Bar 
Association, New . 
York, Longman, 
1986, pp 335-341. 

” Gareth Locksley, 
Technology and 
Capital and Class, 27, 
Winter, 1986, p 89. 

” ibid, p 91. 

‘library’ systems to a closed system of private ownership.” The second 
feature, however, is the application of the principle of indefinite repeti- 
tion that emerges from the technologies of mass production to the tech- 
nologies of production and reproduction of information. 

The copyright is one of the central mechanisms used by both capitalist 
and state capitalist systems to try to reconcile the contradiction between 
these two principles. It is a drive to signature which seeks to limit repeti- 
tion - that is, to limit the potentially infinite iterability of writing and of 
all its technological extensions. But this is doubly impossible. In the 
first place, there is the impossibility of enforcing an artificial scarcity 
imposed on a technology of proliferation. In the second place, there is 
the practical and philosophical impossibility of separating ideas from 
expressions in such a way that private ownership would be extended 
only to the latter without touching the expressed ideas. Copyright doc- 
trine relating to computer software dramatises the contradictions 
involved in the attempt to privatise information, and indeed it demon- 
strates the impossibility of any coherent doctrine of private property in 
intellectual productions. 

S^C - a new quarterly Journal 

■ PILOT ISSUE (Spring 1987): Subscriptions: 4 Issues for 

Star Wars Is already working £20/US$35 individual, 

Humphrey Jennings' Pandaemonlum £35/US$55 Institutional, 

Community radio In 1980s Britain 
The scientist as guru 
Female foeticide in India 

■ISSUE NO. 1 (Summer 1987): 
'Choice' in childbirth 
Home video technology 
Making chips with dust-free poison 
Socially useful production 
Oranta on science 

single copy £6/US$9 

26 Freegrove Road 
London N7 9RQ 
Credit cards (24 hours): 
01-609 5646 



' The concept of the 
double hermeneutic 
derives from Fredric 
Jameson, The Political 
Unconsciouty Ithaca, 
Cornell University 
Press, 1981, especially 
the final chapter. 

THE COMPUTER IS more than an object; it is also an icon and 
a metaphor that suggests new ways of thinking about ourselves and our 
environment, new ways of constructing images of what it means to be 
human and to live in a humanoid world. Cybernetic systems include an 
entire array of machines and apparatuses that exhibit computational 
power. Such systems contain a dynamic, even if limited, quotient of in- 
telligence. Telephone networks, communication satellites, radar sys- 
tems, programmable laser videodiscs, robots, biogenetically engineered 
cells, rocket guidance systems, videotex networks - all exhibit a capacity 
to process information and execute actions. They are all ‘cybernetic’ in 
that they are self-regulating mechanisms or systems within predefined 
limits and in relation to predefined tasks. Just as the camera has come to 
symbolise the entirety of the photographic and cinematic processes, the 
computer has come to symbolise the entire spectrum of networks, 
systems and devices that exemplify cybernetic or ‘automated but in- 
telligent’ behaviour. 

This article traverses a field of inquiry which Walter Benjamin has 
crossed before me, most notably in his 1936 es.say, ‘The Work of Art in 
the Age of Mechanical Reproduction’. My intention, in fact, is to carry 
Benjamin’s inquiry forward and to ask how cybernetic systems, sym- 
bolised by the computer, represent a set of transformations in our con- 
ception of and relation to self and reality of a magnitude commensurate 
with the transformations in the conception of and relation to self and 
reality wrought by mechanical reproduction and symbolised by the 
camera. This intention necessarily encounters the dilemma of a pro- 
found ambivalence directed toward that which constitutes our im- 
aginary Other, in this case not a mothering parent but those systems of 
artificial intelligence I have set out to examine here. Such ambivalence 
certainly permeates Benjamin’s essay and is, at best, dialectical, at 
worst, simply contradictory. Put more positively, those systems against 
which we test and measure the boundaries of our own identity require 
subjection to a double hermeneutic of suspicion and revelation in which 
we must acknowledge the negative, currently dominant tendency to- 
ward control, and the positive, more latent potential toward collectiv- 
ity. ' It will be in terms of law that the dominance of control over collec- 
tivity can be most vividly analysed. 

In summary, what I want to do is recall a few of the salient points in 

Benjamin’s original essay, contrast characteristics of cybernetic systems 
with those of mechanical reproduction, establish a central metaphor 
with which to understand these cybernetic systems, and then ask how 
this metaphor acquires the force of the real; how different institutions 
legitimate their'practices, recalibrate their rationale, and modulate their 
image in fight of this metaphor. In particular, I want to ask how the 
preoccupations of a cybernetic imagination have gained institutional 
legitimacy in areas such as the law. In this case, like others, a tension can 
be seen to exist between the liberating potential of the cybernetic im- 
agination and the ideological tendency to preserve the existing form of 
social relations. I will focus on the work of culture - its processes, opera- 
tions and procedures - and I will assume that culture is of the essence: I 
include within it texts and practices, art and actions that give concrete 
embodiment to the relation we have to existing conditions to a dominant 
mode of production, and the various relations of production it sustains. 
Language, discourse and messages are central. Their style and rhetoric 
basic. Around each ‘fact’ and every ‘datum’, all realities and evidence, 
everything ‘out there’, a persuasive, affective tissue of discourse accrues. 
It is in and through this signifying tissue, arranged in discursive forma- 
tions and institutional arenas, that struggle takes place and semiosis 

Mechanical Reproduction and Film Culture 

Benjamin argues for correspondences among three types of changes; in 
the economic mode of production, in the nature of art and in categories 
of perception. At the base of industrial society lies the assembly line and 
mass production. Technological innovation allows these processes to ex- 
tend into the domain of art, separating off from its traditional ritual (or 
‘cult’) value a new and distinct market (or ‘exhibition’) value. The 
transformation also strips art of its ‘aura’ by which Benjamin means its 
authenticity, its attachment to the domain of tradition: 

The authenticity of a thing is the essence of all that is transmissible from its 
beginning, ranging from its substantive duration to its testimony to the 
history which it has experienced. ^ 

The aura of an object compels attention. Whether a work of art or 
natural landscape, we confront it in one place and only one place. We 
discover its use value in the exercise of ritual, in that place, with that ob- 
ject, or in the contemplation of the object for its uniqueness. The object 
in possession of aura, natural or historical, inanimate or human, engages 
us as if it had ‘the power to look back in return’.’ 

One thing mechanical reproduction cannot, by definition, reproduce 
is authenticity. This is at the heart of the change it effects in the work of 
art. ‘Mechanical reproduction emancipates the work of art from its 
parasitical dependence on ritual’ (p 224). The former basis in ritual 

’ Waller Benjamin, ‘The 
Work of Art in the Age 
of Mechanical 
Reproduction’, in 
translated by Harry 
Zohn, New York, 
Schocken Books, 1969, 
p 221. (Further page 
references from this 
essay will be given in 
the text.) 

’ Walter Benjamin, 
Schriften, 2 vols, 
Frankfurt, Suhrkamp 
Vertag, 1955, 1, p 461. 
Translated in Fredric 
Jameson, Marxism 
and Form, Princeton, 
Princeton University 
Press, 1971, p 77. 

yields to a new basis for art in politics, particularly, for Benjamin, the 
politics of the masses and mass movements, where fascism represents an 
ever-present danger. The possibilities for thoroughgoing emancipation 
are held in check by the economic system surrounding the means of 
mechanical reproduction, especially in film where ‘illusion-promoting 
spectacles and dubious speculations’ {p 232) deflect us from the 
camera’s ability to introduce us to ‘unconscious optics’ that reveal those 
forms of interaction our eyes neglect: 

The act of reaching for a lighter or a spoon is familiar routine, yet we hardly 
know what goes on between hand and metal, not to tnention how this fluc- 
tuates with our moods. Here the camera intervenes with the resources of its 
lowerings and liftings, its interruptions and isolations, its extensions and ac- 
celerations, its enlargements and reductions, (p 237) 

Objects without aura substitute mystique. In a remarkably prescient 
passage, relegated to a footnote, Benjamin elaborates on how political 
practice opens the way for a strange transformation of the actor when 
democracies encounter the crisis of fascism. Mechanical reproduction 
allows the actor an unlimited public rather than the delimited one of the 
stage or, for the politician. Parliament. ‘Though their tasks may be dif- 
ferent, the change affects equally the actor and the ruler This results 

in a new selection, a selection before the equipment (of mechanical 
reproduction) from which the star and the dictator emerge victoriously’ 
(P 247). 

Alterations like the replacement of aura with mystique coincide with 
the third major change posited by Benjamin, change in categories of 
perception. The question of whether film or photography is an art is 
here secondary to the question of whether art itself has not been radical- 
ly transformed in form and function. A radical change in the nature of 
art implies that our very ways of seeing the world have also changed: 
‘During long periods of history, the mode of human sense perception 
changes with humanity’s entire mode of existence’ (p 222). 

Mechanical reproduction makes copies of visible objects, like pain- 
tings, mountain ranges, even human beings, which until then had been 
thought of as unique and irreplaceable. It brings the upheavals of the in- 
dustrial revolution to a culmination. The ubiquitous copy also serves as 
an externalised manifestation of the work of industrial capitalism itself. 
It paves the way for seeing, and recognising, the nature and extent of the 
very changes mechanical reproduction itself produces. 

What element of film most strongly testifies to this new form of 
machine-age perception? For Benjamin it is that element which best 
achieves what Dadaism had aspired to: ‘changes of place and focus 
which periodically assail the spectator’. Film achieves these changes 
through montage, or editing. Montage rips things from their original 
place in an assigned sequence and reassembles them in ever changing 
combinations that make the contemplation invited by a painting impos- 
sible. Montage multiplies the potential of collage to couple two realities 


on a single plane that apparently does not suit them into the juxta- 
position of an infinite series of realities. As Georges Bataille proclaimed^ 
‘Transgression does not negate an interdiction, it transcends and com- 
pletes it.’ In this spirit, montage transcends and completes the project of 
the Dadaists in their conscious determination to strip aura from the 
work of aft and of the early French ethnographers who delighted in the 
strange juxtapositions of artifacts h"om different cultures. 

Montage has a liberating potential, prying art away from ritual and 
toward the arena of political engagement. Montage gives back to the 
worker a view of the world as malleable. Benjamin writes, 

Man’s need to expose himself to shock effects is his adjustment to the changes 
threatening him. The film corresponds to profound changes in the appercep- 
tive apparatus - changes that are experienced on an individual scale by the 
man in the street in big-city traffic, on a historical scale by every present-day 
citizen, (p 250) 

By close-ups of the things around us, by focusing on hidden details of 
familiar objects, by exploring cotnmonplace tnilieus under the ingenious 
guidance of the camera, the film, on the one hand, extends our contprehen- 
sion of the necessities which ride our lives; on the other hand, it manages to 
assure us of an immense and unexpected field of action. Our taverns and our 
metropolitan streets, our offices and furnished rooms, our railroad stations 
and our factories appeared to have us locked up hopelessly. Then came the 
film and burst this prison-world asunder by the dynamite of the tenth of a se- 
cond, so that now, in the midst of its far-flung ruins and debris, we calmly 
and adventurously go travelling, (p 236) 

Mechanical reproduction involves the appropriation of an original, 
although with film even the notion of an original fades; that which is 
filmed has been organised in order to be filmed. This process of ap- 
propriation engenders a vocabulary: the ‘take’ or ‘camera shot’ used to 
‘shoot’ a scene where both stopping a take and editing are called a ‘cut’. 
The violent re-ordering of the physical world and its meanings provides 
the shock effects Benjamin finds necessary if we are to come to terms 
with the age of mechanical reproduction. The explosive, violent poten- 
tial described by Benjamin and celebrated by Brecht is what the domi- 
nant cinema must muffle, defuse and contain. And what explosive 
potential can be located in the computer and its cybernetic systems for 
the elimination of drudgery and toil, for the promotion of collectivity 
and affinity, for interconnectedness, systemic networking and shared 
decision-making, this, too, must be defused and contained by the in- 
dustries of information which localise, condense and consolidate this 
potential deraocratisation of power into hierarchies of control. 

‘Montage - the connecting of dissimilars to shock an audience into 
insight - becomes for Benjamin a major principle to artistic production 
in a technological age.’"' Developing new ways of seeing to the point 
where they become habitual is not ideological for Benjamin but trans- 
formative. They are not the habits of old ways but new; they are skills 

■* Terry Eagleton, 
Marxism and Literary 
Criticism, Berkeley, 
University of 
California Press, 1976, 
p 63. 

^ This quote is from 
James Clifford, ‘On 
Surrealism’, in 
Comparative Studies in 
Society and History vol 
23 no 4, October 
1981, pp 559-564, 
where he offers an 
excellent description 
of the confluences 
between surrealism 
and certain tendencies 
within early 
ethnography in 1920s 

which are difTicult to acquire precisely because they are in opposition to 
ideology. The tasks before us ‘at the turning points of history’ cannot be 
met by contemplation. ‘They are mastered gradually by habit, under the 
guidance of tactile appropriation’ (p 240). The shocks needed in order to 
adjust to threatening changes may be co-opted by the spectacles a 
culture industry provides. For Benjamin the only recourse is to those 
skills he himself adopted: the new habits of a sensibility trained to 
disassemble and reconstruct reality, of a writing style intended to relieve 
idlers of their convictions, of a working class trained not only to produce 
and reproduce the existing relations of production but to reproduce 
those very relations in a new, liberating form. ‘To see culture and its 
norms - beauty, truth, reality - as artificial arrangements, susceptible to 
detached analysis and comparison with other possible dispositions’ 
becomes the vantage point not only of the surrealist but the 

The process of adopting new ways of seeing that consequently pro- 
pose new forms of social organisation becomes a paradoxical, or dialec- 
tical, process when the transformations that spawn new habits, new vi- 
sion, are themselves engendered and substantially recuperated by the 
existing form of social organisation which they contain the potential to 
overcome. But the process goes forward all the same. It does so less in 
terms of a culture of mechanical reproduction, which has reached a 
point similar to that of a tradition rooted in Benjamin’s time, than in 
terms of a culture of electronic dissemination and computation. 

We might then ask in what ways is our ‘sense of reality’ being adjusted 
by new means of electronic computation and digital communication? 
Do these technological changes introduce new forms of culture into the 
relations of production at the same time as the ‘shock of the new’ helps 
emancipate us from the acceptance of social relations and cultural forms 
as natural, obvious or timeless? The distinction between an industrial 
capitalism, even in its ‘late’ phase of monopoly concentration, and an in- 
formation society that does not ‘produce’ so much as ‘process’ its basic 
forms of economic resource has become an increasingly familiar distinc- 
tion for us. Have cybernetic systems brought about changes in our 
perception of the world that hold liberating potential? Is it conceivable, 
for example, that contemporary transformations in the economic struc- 
ture of capitalism, attended by technological change, institute a less in- 
dividuated, more communal form of perception similar to that which 
was attendant upon face-to-face ritual and aura but which is now 
mediated by anonymous circuitry and the simulation of direct en- 
counter? Does montage now have its equivalent in interactive simula- 
tions and simulated interactions experienced according to predefined 
constraints? Does the work of art in the age of postmodernism lead, at 
least potentially, to apperceptions of the ‘deep structure’ of post- 
industrial society comparable to the apperceptive discoveries occasioned 
by mechanical reproduction in the age of industrial capitalism? 

Cybernetic Systems and Electronic Culture 


We can put Benjamin’s arguments, summarised cursorily here, in 
another perspective by highlighting some of the characteristics 
associated with early, entrepreneurial capitalism, monopoly capitalism 
and multinational or post-industrial capitalism: 







steam and locomotive 

electricity and petro- 

microelectronics and 


chemical power 

nuclear energy 

property rights 

corporate rights 

copyright and patents 

nature as Other/con- 

aliens as Other/con- 

knowledge as Other/con- 

quest of nature 

quest of Third World 

quest of intelligence 




working class vanguard 

consumer group 

aflinity group vanguard 




- contamination by 

- contamination by an 

- deficiency of self (col- 


aberrant self 

lapse of immune system 
that distinguishes self 
from environment) 

- isolation of self from 

- isolation of aberrant 

- isolation of self by 


tissue from self 

artificial life support 


- vulnerability to 

- vulnerability to 

- vulnerahility to 


systemic collapse 

invasive agents 

- heightened 

- heightened sense of 

- heightened 





po stmodemism 




mechanical repro- 

instantaneous broad- 

logico-iconic simulations 



reproducible instances 

ubiquitous occurrences 

processes of absorption 
and feedback 

the copy 

the event 

the chip (and VDT 

subtext of possession 

subtext of mediation 

subtext of control 

image and represen- 

collage and Juxta- 




Simulacra introduce the key question of how the control of informa- 
tion moves toward control of sensory experience, interpretation, in- 
telligence and knowledge. The power of the simulation moves to the 


^ See, for example, the 
essays in Part III, 
‘Form and Pathology 
in Relationship’ of 
Gregory Bateson, 
S(eps loan Ecology of 
New York, 
Ballantine, 1972, 
where this phrase is 
introduced and 
applied to various 

heart of the cybernetic matter. It posits the simulation as an imaginary 
Other which serves as the measure of our own identity and, in doing so, 
prompts the same form of intense ambivalence that the mothering 
parent once did: a guarantee of identity based on what can never be 
made part of oneself. In early capitalism the human was defined in rela- 
tion to an animal world that evoked fascination and attraction, repulsion 
and resentment. The human animal was similar to but different from all 
other animals. In monopoly capitalism the human was defined in rela- 
tion to a machine world that evoked its own distinctive blend of am- 
bivalences. The human machine was similar to but different from all 
other machines. In post-industrial capitalism the human is defined in 
relation to cybernetic systems - computers, bio-genetically engineered 
organisms, eco-systems, expert systems, robots, androids and cyborgs - 
all of which evoke those forms of ambivalence reserved for the Other 
that is the measure of ourselves. The human cyborg is similar to but dif- 
ferent from all other cyborgs. Through these transformations questions 
of difference persist. Human identity remains at stake, subject to 
change, vulnerable to challenge and modification as the very metaphors 
prompted by the imaginary Others that give it form themselves change. 
The metaphor that’s meant (that’s taken as real) becomes the simula- 
tion. The simulation displaces any antecedent reality, any aura, any 
referent to history. Frames collapse. What had been fixed comes 
unhinged. New identities, ambivalently adopted, prevail. 

The very concept of a text, whether unique or one of myriad copies, 
for example, underpins almost all discussion of cultural forms including 
film, photography, and their analogue in an age of electronic com- 
munication, television (where the idea of ‘flow’ becomes an impor- 
tant amendment). But in cybernetic systems, the concept of ‘text’ itself 
undergoes substantial slippage. Although a textual element can still be 
isolated, computer-based systems are primarily interactive rather than 
one-way, open-ended rather than fixed. Dialogue, regulated and dis- 
seminated by digital computation, deemphasises authorship in favour of 
‘messages-in-circuit’* that take fixed but effervescent, continually var- 
iable form. The link between message and substrate is loosened: words 
on a printed page are irradicable; text on a VDT (video display terminal) 
is readily altered. The text conveys the sense of being addressed to us. 
The message-in-circuit is both addressed to and addressable by us; the 
mode is fundamentally interactive, or dialogic. That which is most tex- 
tual in nature -the fixed, read-only-memory (ROM) and software 
programs - no longer addresses us. Such texts are machine addressable. 
They direct those operational procedures that ultimately give the im- 
pression that the computer responds personally to us, simulating the 
processes of conversation or of interaction with another intelligence to 
effect a desired outcome. Like face-to-face encounter, cybernetic sys- 
tems offer (and demand) almost immediate response. This is a major 
part of their hazard in the workplace and their fascination outside it. 
The temporal flow and once-only quality of face-to-face encounter 

Cyborg: the human brain as circuit board. (Illustration courtesy of Program magazine.) 

becomes embedded within a system ready to restore, alter, modify or 
transform any given moment to us at any time. Cybernetic interactions 
can become intensely demanding, more so than we might imagine from 
our experience with texts, even powerfully engaging ones. Reactions 
must be almost instantaneous, grooved into eye and finger reflexes until 
they are automatic. This is the bane of the ‘automated workplace’ and 
the joy of the video game. Experienced video game players describe 
their play as an interactive ritual that becomes totally self-absorbing. As 
David, a lawyer in his mid-30s interviewed by Sherry Turkic, puts it. 

. . . At the risk of sounding, uh, ridiculous, if you will, it’s almost a Zen type 
of thing. . . . When I can direct myself totally but not feel directed at all. 


^ Quoted in Sherry 
Turkic, The Second 
Self: Computer and the 
Human Spirit, New 
York, Simoii and 
Schuster, 1984, p 86. 

You’re totally absorbed and it’s all happening there. . . . You either get 
through this little maze so that the creature doesn’t swallow you up or you 
don’t. And if you can focus your attention on that, and if you can really 
learn what you’re supposed to do, then you really are in relationship with the 
game. ’’ 

® Steven J Heims, JoAn 
von Neuman and 
Norbert Wiener: From 
Mathematics to the 
Technologies of Life and 
Death, Cambridge, 
Mass, MIT Press, 

1 980, describes how 
research on anti- 
aircraft guidance 
systems led Julian 
Bigelow and hinrbert 
Wiener to develop a 
mathematical theory 
‘for predicting the 
future as best one can 
on the basis of 
incomplete information 
about the past’ (p 183). 
For an overview of the 
history of cybernetic 
theory and cognitive 
psychology in the 
context of its military- 
industrial origins, see 
Paul N Edwards, 
‘Formalized Warfare’, 
unpublished ms (1984), 
History of 
Program, University of 
California, Santa Cruz. 

^ Jean Baudrillard, ‘The 
Implosion of Aieaning 
in the Media and the 
Implosion of the Social 
in the Masses’, in 
Kathleen Woodward 
(ed). The Myths of 
Information, Madison, 
Coda Press, 1980, p 

The enhanced ability to test the environment, which Benjamin 
celebrated in film (‘The camera director in the studio occupies a place 
identical with that of the examiner during aptitude tests,’ p 246) certain- 
ly continues with cybernetic communication.® The computer’s dialogic 
mode carries the art of the ‘what if even further than the camera eye has 
done, extending beyond the ‘what if I could see more than the human 
eye can see’ to ‘what if I can render palpable those possible transforma- 
tions of existing states that the individual mind can scarcely 

If mechanical reproduction centres on the question of reproducibility 
and renders authenticity and the original problematic, cybernetic 
simulation renders experience, and the real itself, problematic. Instead 
of reproducing, and altering, our relation to an original work, cybernetic 
communication simulates, and alters, our relation to our environment 
and mind. As Jean Baudrillard argues, ‘Instead of facilitating com- 
munication, it (information, the message-in-circuit) exhausts itself in the 
staging of communication . . . this is the gigantic simulation process with 
which we are familiar.’’ Instead of a representation of social practices 
recoded into the conventions and signs of another language or sign- 
system, like the cinema, we encounter simulacra that represent a new 
form of social practice in their own right and re-present nothing. The 
photographic image, as Roland Barthes proposed, suggests ‘having been 
there’ of what it represents, of what is present-in-absentia. The com- 
puter simulation suggests only a ‘being here’ and ‘having come from 
nowhere’ of what it presents, drawing on those genetic-like algorithms 
that allow it to bring its simulation into existence, sui generis. Among 
other things, computer systems simulate the dialogical and other 
qualities of life itself. The individual becomes nothing but an ahistorical 
position within a chain of discourse marked exhaustively by those 
shifters that place him or her within speech acts (‘I’, ‘here’, ‘now’, ‘you’, 
‘there’, ‘then’). In face-to-face encounter this ‘I’ all speakers share can be 
inflected to represent some part of the self not caught by words. To res- 
pond to the query, ‘How are you?’ by saying ‘Not too bad’, rather than 
‘Fine’ suggests something about a particular state of mind or style of ex- 
pression and opens onto the domains of feeling and empathy. What can- 
not be represented in language directly (the bodily, living ‘me’ that 
writes or utters words) can significantly inflect speech, and dialogue, 
despite its enforced exclusion from any literal representation. 

In cybernetic systems, though, ‘I’ and ‘you’ are strictly relational pro- 
positions attached to no substantive body, no living individuality. In 
place of human intersubjectivity we discover a systems interface, a 

boundary between cyborgs that selectively passes information but 
without introducing questions of consciousness or the unconscious^ 
desire or will, empathy or conscience, saved in simulated forms. 

Even exceptions like ELIZA, a program designed to simulate a 
therapeutic encounter, prove the rule. ‘I’ and ‘you’ function as partners 
in therapy only as long as the predefined boundaries are observed. As 
Sherry Turkle notes, if you introduce the word ‘mother’ into your ex- 
change, and then say, ‘Let’s discuss paths toward nuclear disarmament,’ 
ELIZA might well offer the nonsense reply, ‘Why are you telling me 
that your mother makes paths toward nuclear disarmament?’'® Simula- 
tions like these may bring with them the shock of recognising the reifica- 
tion of a fundamental social process, but they also position us squarely 
within a realm of communication and exchange cleanly evacuated of the 
intersubjective complexities of direct encounter. Cybernetic systems 
give form, external expression, to processes of mind (through messages- 
in-circuit) such that the very ground of social cohesion and conscious- 
ness becomes mediated through a computational apparatus. Cybernetic 
interaction achieves with an other (an intelligent apparatus) the simul- 
ation of social process itself. 

Cybernetic dialogue may offer freedom from many of the apparent 
risks inherent in direct encounter; it offers the illusion of control. This 
use of intelligence provides a lure that seems to be much more attractive 
to men than women. At first there may seem to be a gain, particularly 
regarding the question of the look or gaze. Looking is an intensely charg- 
ed act, one significantly neglected by Benjamin, but stressed in recent 
feminist critiques of dominant Hollywood cinema. There looking is pos- 
ed as a primarily masculine act and ‘to-be-looked-at-ness’ a feminine 
state, reinforced, in the cinema, by the camera’s own voyeuristic gaze, 
editing patterns that prompt identification with masculine activism and 
feminine passivity, and a star system that institutionalises these uses of 
the look through an iconography of the physical body." This entire 
issue becomes circumvented in cybernetic systems that simulate dia- 
logic interaction, or face-to-face encounter, but exclude not only the 
physical self or its visual representation but also the cinematic apparatus 
that may place the representation of sexual difference within a male- 
dominant hierarchy. 

Correct in so far as it goes, the case for the circumvention of the sexist 
coding of the gaze overlooks another form of hierarchical sexual coding 
that revolves around the question of whether a fascination with 
cybernetic systems is not itself a gender-related (i.e., a primarily 
masculine) phenomenon (excluding from consideration an even more 
obvious gender coding that gives almost all video games, for example, a 
strong aura of aggressive, militaristic activity). The questions that we 
pose about the sexist nature of the gaze within the cinematic text and the 
implications this has for the position we occupy in relation to such texts, 
may not be wholly excluded so much as displaced. A (predominantly 
masculine) fascination with the control of simulated interactions 

'® Sherry Turkic, op dt, 

' ' See Laura Mulvey, 
‘Visual Pleasure and 
Narrative Cinema’, 
Screen Autumn 1975, 
vol 16 no 3, pp 6-18. 


Paul N Edwards, op 
cit, p 59. 

replaces a (predominantly masculine) fascination with the to-be-looked- 
at-ness of a projected image. Simulated intersubjectivity as a product of 
automated but intelligent systems invokes its own peculiar psycho- 
dynamic. Mechanical reproduction issues an invitation to the fetishist: a 
special relationship to the images of actors or politicians in place of any 
more direct association. The fetish object - the image of the other that 
takes the place of the other - becomes the centre of attention while 
fetishistic viewers look on from their anonymous and voyeuristic, 
seeing-but-unseen sanctuary in the audience. But the output of com- 
putational systems stresses simulation, interaction and process itself. 
Engagement with this process becomes the object of fetishisation rather 
than representations whose own status as produced objects has been 
masked. Cybernetic interaction emphasises the fetishist rather than the 
fetish object: instead of a taxonomy of stars we find a galaxy of computer 
freaks. The consequence of systems without aura, systems that replace 
direct encounter and realise otherwise inconceivable projections and 
possibilities, is a fetishism of such systems and processes of control 
themselves. Fascination resides in the subordination of human volition 
to the operating constraints of the larger system. We can talk to a system 
whose responsiveness grants us an awesome feeling of power and con- 
trol. But as Paul Edwards observes, ‘Though individuals . . . certainly 
make decisions and set goals, as links in the chain of command they are 
allowed no choices regarding the ultimate purposes and values of the 
system. Their “choices” are . . . always the permutations and combina- 
tions of a predefined set.’*^ 

The desire to exercise a sense of control over a complex but predefined 
logical universe replaces the desire to view the image of an Other over 
which the viewer can imagine himself to have a measure of control. The 
explosive power of the dynamite of the tenth of a second extolled by 
Benjamin is contained within the channels of a psychopathology that 
leave exempt from apperception, or control, the mechanisms that place 
ultimate control on the side of the cinematic apparatus or cybernetic 
system. These mechanisms - the relay of gazes between camera, 
characters and viewer, the absorption into a simulacrum with complex 
problems and eloquent solutions - are the ground upon which engage- 
ment occurs and are not addressable within the constraints of the system 
itself. It is here, at this point, that dynamite must be applied. 

This is even more difficult with computers and cybernetics than with 
cameras and the cinema. Benjamin himself noted how strenuous a task it 
is in film to mask the means of production, to keep the camera and its 
supporting paraphernalia and crew from intruding upon the fiction. Ex- 
posure of this other scene, the one behind the camera, is a constant 
hazard and carries the risk of shattering the suspension of disbelief. 
Only those alignments between camera and spectator that preserve the 
illusion of a fictional world without cameras, lights, directors, studio 
sets and so on are acceptable. Benjamin comments, perhaps with more 
of a surrealist’s delight in strange juxtapositions than a marxist’s, ‘The 

equipment-free aspect of reality here (in films) has become the height of 
artificej the sight of immediate reality has become an orchid in the land 
of technology’ (p 233). 

Within the contemporary prison-house of language, in Fredric Jame- 
son’s apt phrase, the orchid of immediate reality, like the mechanical 
bird seen at the end of Blue Velvet, appears to have been placed per- 
manently under glass, but for Benjamin neither the process by which an 
illusionistic world is produced nor the narrative strategies associated 
with it receive extended consideration. For him, the reminders of the 
productive process were readily apparent, not least through the stren- 
uous efforts needed to mask them. The ‘other scene’ where fantasies and 
fictions actually become conceptually and mechanically produced may 
be repressed but it is not obliterated. If not immediately visible, it lurks 
just oiit of sight in the off-screen space where the extension of a fictional 
world somewhere collides with the world of the camera apparatus in one 
dimension and the world of the viewer in another. It retains the poten- 
tial to intrude at every ‘cut’ or editj it threatens to reveal itself in every 
lurch of implausibility or sleight-of-hand with which a narrative at- 
tempts to achieve the sense of an ending. 

With cybernetic systems, this other scene from which complex rule- 
governed universes actually get produced recedes further from sight. 
The governing procedures no longer address us in order to elicit a 
suspension of disbelief; they address the cybernetic system, the micro- 
processor of the computer, in order to absorb us into their operation. 
The other scene has vanished into logic circuits and memory chips, into 
‘machine language’ and interface cards. The chip replaces the copy. Just 
as the mechanical reproduction of copies revealed the power of in- 
dustrial capitalism to reorganise and reassemble the world around us, 
rendering it as commodity art, the automated intelligence of chips 
reveals the power of post-industrial capitalism to simulate and replace 
the world around us, rendering not only that exterior realm but also in- 
terior ones of consciousness, intelligence, thought and intersubjectivity 
as commodity experience. 

The chip is pure surface, pure simulation of thought. Its material sur- 
face is its meaning - without history, without depth, without aura, affect 
or feeling. The copy reproduces the world, the chip simulates it. It is the 
difference between being able to remake the world and being able to ef- 
face it. The micro-electronic chip draws us into a realm, a design for liv- 
ing, that fosters a fetishised relationship with the simulation as a new 
reality all its own based on the capacity to control, within the domain of 
the simulation, what had once eluded control beyond it. The orchids of 
immediate reality that Benjamin was wont to admire have become the 
paper flowers of the cybernetic simulation. 

Electronic simulation instead of mechanical reproduction. Fetishistic 
addiction to a process of logical simulation rather than a fascination with 
a fetishised object of desire. Desire for the dialogic or interactive and the 
illusion of control versus desire for the fixed but unattainable and the il- 

lusion of possession. Narrative and realism draw us into relations of 
identification with the actions and qualities of characters. Emulation is 
possible, as well as self-enhancement. Aesthetic pleasure allows for a 
revision of the world from which a work of art arises. Reinforcing what 
is or proposing what might be, the work of art remains susceptible to a 
double hermeneutic of suspicion and revelation. Mechanical reproduc- 
tion changes the terms decidedly, but the metonymic or indexical rela- 
tionship between representational art and the social world to which it 
refers remains a fundamental consideration. 

By contrast, cybernetic simulations offer the possibility of completely 
replacing any direct connection with the experiential realm beyond their 
bounds. Like the cinema, this project, too, has its origins in the expan- 
sion of nineteenth century industrialism. The emblematic precursors of 
the cyborg - the machine as self-regulating system - were those animate, 
self-regulating systems that offered a source of enchantment even 
museums could not equal: the zoo and the botanical garden. 

At the opening of the first large-scale fair or exhibition, the Great Ex- 
hibition of 1851, Queen Victoria spoke of ‘the greatest day in our history 
(when) the whole world of nature and art was collected at the call of the 
queen of cities’. Those permanent exhibitions - the zoo and botanical 
garden - introduced a new form of vicarious experience quite distinct 
from the aesthetic experience of original art or mechanically reproduced 
copies. The zoo brings back alive evidence of a world we could not 
otherwise know, now under apparent control. It offers experience at a 
remove that is fundamentally different as a result of having been 
uprooted from its original context. The indifferent, unthreatened and 
unthreatening gaze of captive animals provides eloquent testimony to 
the difference between the zoo and the natural habitat to which it refers. 
The difference in the significance of what appears to be the same thing, 
the gaze, indicates that the change in context has introduced a new 
system of meanings, a new discourse or language. 

Instead of the shocks of montage that offer a ‘true means of exercise’ 
appropriate to the ‘profound changes in the apperceptive apparatus’ 
under industrial capitalism, the zoo and botanical garden exhibit a 
predefined, self-regulating world with no reality outside of its own 
boundaries. These worlds may then become the limit of our understan- I 
ding of those worlds to which they refer but of which we seldom have 
direct knowledge. ‘Wildlife’ or ‘the African savannah’ is its simulation 
inside the zoo or garden or diorama. Absorption with these simulacra 
and the sense of control they afford may be an alternative means of exer- 
cise appropriate to the apperceptive changes required by a service and 
information economy. 

Computer-based systems extend the possibilities inherent in the zoo 
and garden much further. The ideal simulation would be a perfect 
replica, now controlled by whoever controls the algorithms of 
simulation - a state imaginatively rendered in films like The Stepford 
Wives or Bladerunner and apparently already achieved in relation to cer- 

tain biogenetically engineered micro-organisms. Who designs and con- 
trols these greater systems and for what purpose becomes a question of 
central importance. 

The Cybernetic Metaphor: Transformations of Self and Reality 

The problems of tracking anti-aircraft weapons against extremely fast 
targets prompted the research and development of intelligent mechan- 
isms capable of predicting future states or positions far faster than the 
human brain could do. The main priorities were speed, efficiency 
and reliability, i.e., fast-acting, error-free systems. ENIAC (Electronic 
Numerical Integrator and Computer), the first high-powered digital 
computer, was designed to address precisely this problem by perform- 
ing ballistic computations at enormous speed and allowing the outcome 
to be translated into adjustments in the firing trajectory of anti-aircraft 

The men (sic) who assembled to solve problems of this order and who 
formalised their approach into the research paradigms of information 
theory and cognitive psychology through the Macy Foundation Con- 
ferences, represent a who’s who of cybernetics: John von Neuman, 
Oswald Weblen, Vannevar Bush, Norbert Wiener, Warren McCulloch, 
Gregory Bateson and Claude Shannon, among others.” Such research 
ushers in the central metaphors of the cybernetic imagination: not only 
the human as an automated but intelligent system, but also automated, 
intelligent systems as human, not only the simulation of reality but the 
reality of the simulation. These metaphors take form around the ques- 
tion, the still unanswered question, put by John Stroud at the Sixth 
Macy Conference; 

IFe know as much as possible about how the associated gear bringing the in- 
formation to the tracker [of an anti-aircraft gun] operates and how all the 
gear from the tracker to the gun operates. So we have the human operator 
surrounded on both sides by very precisely known mechanisms atid the ques- 
tion comes up, ‘What kind of machine have we placed in the middle?"^'^ 

This question of ‘the machine in the middle’ and the simulation as reali- 
ty dovetails with Jean Baudrillard’s recent suggestion that the staging 
powers of simulation establish a hyperreality we only half accept but 
seldom refute, ‘Hyperreality of communication of meaning: by dint of 
being more real than the real itself, reality is destroyed.’” 

Such metaphors, then, become more than a discovery of similarity, 
they ultimately propose an identity. Norbert Wiener’s term ‘cyborg’ 
(cybernetic organism) encapsulates the new identity which, instead of 
seeing humans reduced to automata, sees simulacra which encompass 
the human elevated to the organic. Consequently, the human cognitive 
apparatus (itself a hypothetical construct patterned after the cybernetic 

* ^ See, for example, 

Paul N Edwards, ibid, 
for a more detailed 
account of this 
synergism between 
the development of 
cybernetics and 
military needs. For a 
cybernetic theory of 
alcoholism and 
schizophrenia, see 
Gregory Bateson, op 
cit, and Watzlawick, 
Beavin and Jackson’s 
study of human 
interaction in a 
systems framework in 
Pragmatics oj Human 

’■* John Stroud 
Moments in 
Perception - 
Discussions’, in H 
Van Foersta, et al 
(eds). Cybernetics: 
Circular Causal and 
Feedback Mechanisms 
in Biological and Social 
Systems, Transactions 
of the Sixth Macy 
Conference, New 
York, Josiah Macy 
Foundation, 1949, pp 

” Jean Baudrillard, op 
cit, p 139. 


model of automated intelligence) i^xpected to negotiate the world by 
means of simulation. 

Our cognitive apparatus treats the real as thought it consisted of those 
properties exhibited by simulacra. The real becomes simulation. 
Simulacra, in turn, serve as the mythopoeic impetus for that sense of the 
real we posit beyond the -simulation. A sobering example of what is at 
stake follows from the Reagonomic conceptualisation of war. The 
Strategic Defense Initiative represents a vast Battle of the Cyborgs video 
game where players compete to save the world from nuclear holocaust. 
Reagan’s simulated warfare will turn the electromagnetic force fields of 
’50s science-fiction films that shielded monsters and creatures from the 
arsenal of human destructive power into ploughshares beyond the 
ozone. Star Wars will be the safe sex version of international conflict: 
not one drop of our enemy’s perilous bodily fluids, none of their nuclear 
ejaculations, will come into contact with the free world. 

Reagan’s simulation of war as a replacement for the reality of war does 
not depend entirely on the SDI. We have already seen it at work in the 
invasion of Grenada and the raid on Libya. Each time, we have had the 
evocation of the reality of war: the iconography of heroic fighters, em- 
battled leaders, brave decisions, powerful technology and concerted ef- 
fort rolled into the image of military victory, an image of quick, decisive 
action that defines the ‘American will’. 

These simulacra of war, though, are fought with an imaginary enemy, 
in the Lacanian sense, and in the commonsense meaning of an enemy 

posited within those permutations allowed by a predefined set of 
assumptions and foreign policy options: a Grenadian or Libyan ‘threat’ 
appears on the video screens of America’s political leadership. Long ex- 
perience with the communist menace leads to prompt and sure recogni- 
tion. Ronny pulls the trigger. These simulations lack the full-blown, 
catastrophic consequences of real war, but this does not diminish the 
reality of this particular simulation nor the force with which it is map- 
ped onto a historical ‘reality’ it simultaneously effaces. Individuals find 
their lives irreversibly altered, people are wounded, many die. These in- 
delible punctuation marks across the face of the real, however, fall into 
place according to a discourse empowered to make the metaphoric reali- 
ty of the simulation a basic fact of existence. 

A more complex example of what it means to live not only in the socie- 
ty of the spectacle but also in the society of the simulacrum involves the 
preservation/simulation of life via artificial life-support systems. In such 
an environment, the presence of life hinges on the presence of ‘vital 
signs’. Their manifestation serves as testimony to the otherwise inac- 
cessible presence of life itself, even though life in this state stands in rela- 
tion to the ‘immediate reality’ of life as the zoo stands in relation to 
nature. The important issue here is that the power of cybernetic simula- 
tions prompts a redefinition of such fundamental terms as life and reali- 
ty, just as, for Benjamin, mechanical reproduction alters the very con- 
ception of art and the standards by which we know it. Casting the issue 
in terms of whether existence within the limits of an artificial life- 
support system should be considered ‘life’ obscures the issue in the same 
way that asking whether film and photography are ‘art’ does. In each 
case a presumption is made about a fixed, or ontologicaily given nature 
to life or art, rather than recognising how that very presumption has 
been radically overturned. 

And from preserving life artificially it is a small step to creating life by 
the same means. There is, for example, the case of Baby M. Surrogate 
mothering, as a term, already demonstrates the reality of the simulation: 
the actual mothering agent - the woman who bore the child - becomes a 
surrogate, thought of, not as a mother, but as an incubator or ‘rented 
uterus’ as one of the trial’s medical ‘experts’ called Mary Beth 
Whitehead. The real surrogate mother, the woman who will assume the 
role of mother for a child not borne of her own flesh, becomes the real 
mother, legally and familially. The law upholds the priority of the 
simulation and the power of those who can control this system of 
surrogacy - measured by class and gender, for it is clearly upper-class 
males (Judge Harvey Sorkow and the father, William Stern) who 
mobilised and sanctioned this particular piece of simulation, largely, it 
would seem, given the alternative of adoption, to preserve a very real, 
albeit phantastic preoccupation with a patriarchal blood line. 

I Here we have the simulation of a nuclear family - a denucleated, ar- 
tificial simulation - made and sanctioned as real, bona fide. The trial 
evoked the reality of the prototypical bourgeois family: well-educated, 


Sec Nicos Poulantzas, 
Political PoTver and 
Social Class, London, 
New Left Books, 
1975, pp 211-214. 

socially responsible, emotionally stable and economically solvent, in 
contrast to the lower middle-class Whitehead household. The trial judg- 
ment renders as legal verdict the same moral lesson that Cecil Hep- 
worth’s 1905 film. Rescued by Rover, presents as artistic theme: the pro- 
priety of the dominant class, the menace of an unprincipled, jealous and 
possessive lower class, the crucial importance of narrative donors like 
the faithful Rover and of social agents like the patronising Sorkow and 
the central role of the husband as the patriarch able to preside over the 
constitution and re-constitution of his family. Now replayed as simula- 
tion the morality play takes on a reality of its own. People suffer, 
wounds are inflicted. Lives are irreversibly altered, or even created. 
Baby M is a child conceived as a product to be sold to fill a position 
within the signifying discourse of patriarchy. 

The role of the judge in this case was, of course, crucial to its outcome. 
His centrality signals the importance of the material, discursive strug- 
gles being waged within the realm of the law. Nicos Poulantzas'* argues 
that the juridico-political is the dominant or articulating region in 
ideological struggle today. Law establishes and upholds the conceptual 
frame in which subjects, ‘free and equal’, with ‘rights’ and ‘duties’, 
engage on a playing field made level by legal recourse and due process. 
These fundamental concepts of individuals with the right to enter into 
and withdraw from relations and obligations to others underpin, he 
argues, the work of other ideologically important regions in civil society. 

Whether the juridico-political is truly the fulcrum of ideological con- 
testation or not, it is clearly a central area of conflict and one in which 
some of the basic changes in our conception of the human/computer, 
reality/simulation metaphors get fought out. Re-conceptualisations of 
copyright and patent law, brought on by computer chip design, com- 
puter software, and biogenetic engineering, give evidence of the process 
by which a dominant ideology seeks to preserve itself in the face of 
historical change. 

Conceptual metaphors take on tangible embodiment through discur- 
sive practices and institutional apparatuses. Such practices give a 
metaphor historical weight and ideological power. Tangible embodi- 
ment has always been a conscious goal of the cybernetic imagination 
where abstract concepts become embedded in the logic and circuitry of a 
material substrate deployed to achieve specific forms of result such as a 
computer, an anti-aircraft tracking system or an assembly line robot. 
These material objects, endowed with automated but intelligent capacit- 
ies, enter our culture as, among other things, commodities. As a peculiar 
category of object these cyborgs require clarification of their legal status. 
What proprietary rights pertain to them? Can they be copyrighted, 
patented, protected by trade secrets acts; can they themselves as 
automated but intelligent entities, claim legal rights that had previously 
been reserved for humans or other living things on a model akin to that 
which has been applied to animal research? 

The answers to such questions do not fall from the sky. They are the 


result of struggle, of a clash of forces, and of the efforts, faltering or elo- 
quent, of those whose task it is to make and adjudicate the law. New 
categories of objects do not necessarily gain the protection of patent or 
copyright law. One reason for this is that federal law in the United States 
(where most of my research on this question took place) and the Con- 
stitution both enshrine the right of individuals to private ownership of 
the means of production while also enjoining against undue forms of 
monopoly control. The Constitution states, ‘The Congress shall have 
power ... to promote the progress of science and useful arts, by securing 
for limited times to authors and inventors the exclusive right to their 
respective writings and discoveries.’ Hence the protection ofintellectual 
property (copyright and trademark registration) or industrial and 
technological property (patents) carves out a proprietary niche within 
the broader principle of a ‘free flow’ of ideas and open access to ‘natural’ 
sources of wealth. 

The cybernetic organism, of course, confounds the distinction bet- 
ween intellectual and technological property. Both a computer and a 
biogenetically designed cell ‘may be temporarily or permanently pro- 
grammed to perform many different unrelated tasks’. The cybernetic 
metaphor, of course, allows us to treat the cell and the computer as 
sources of the same problem. As the author of one legal article observed, 
‘A ribosome, like a computer, can carry out any sequence of assembly in- 
structions and can assemble virtually unlimited numbers of different 
organic compounds, including those essential to life, as well as materials 
that have not yet been invented.’’® What legal debates have characteris- 
ed the struggle for proprietary control of these cyborgs? 

Regarding patents, only clearly original, unobvious, practical applica- 
tions of the ‘laws of nature’ are eligible for protection, a principle firmly 
established in the Telephone Cases of 1888 where the Supreme Court 
drew a sharp distinction between electricity itself as non-patentable 
since it was a ‘force of nature’ and the telephone where electricity was 
found, ‘A new, specific condition not found in nature and suited to the 
transmission of vocal or other sounds’. 

Recent cases have carried the issue further, asking whether ‘in- 
telligent systems’ can be protected by patent and, if so, what specific 
elements of such a system are eligible for protection. Generally, and 
perhaps ironically, the US Supreme Court has been more prone to grant 
protection for the fabrication of new life forms, via recombinant DNA 
experiments, than for the development of computer software. In Dia- 
mond V Chakrabatry (1980), the Supreme Court ruled in favour of 
patent protection for Chakrabatry who had developed a new bacterial 
form capable of degrading petroleum compounds for projected use in 
oil-spill clean-ups. In other, earlier cases, the Supreme Court withheld 
patent protection for computer software. In Gottschalk v Benson (1972) 
and in Parker v Flook (1979), the Court held that computer programs 
were merely algorithms, i.e., simple, step-by-step mathematical pro- 
cedures, and as such were closer to basic principles or concepts than to 

' ^ James J Alyrick and 
James A Sprowl, 
‘Patent Law for 
Computers and 
Programmed Life 
Forms’, American Bar 
Association Journal no 
68, August 1982, p 

'® ibid, p 121. Some 
other relevant articles 

Patent Law 
Developments in 
Great Britain and the 
United States’, Boston 
College International 
and Comparative Lavt 
Review no 6, Spring 
1983, pp 563-590; 

‘Can a Computer be 
an Author? Copyright 
Aspects of Artificial 
tainment Law Journal 
no 4, Summer 1982, 
pp 707-747; Peter 

‘Copyright Protection 
for Computer 
Programs in Read 
Only Memory Chips’, 
Hojstra Law Review 
no 11, February 1982, 
pp 329-370; ‘Patents 
on Algorithms, 
Discoveries and 
Scientific Principles’, 
Idea no 24, 1 983, pp 
21-39; S Hewitt, 
‘Protection of Works 
Created by Use of 
Computers’, New Law 
Journal no 1 33, 

March 11, 1983, pp 
235-237; EN 
Kramsky, ‘Video 
Games: Our Legal 
System Grapples with 
a Social 

Journal of the Patent 
Office Society no 64, 
June 1982, pp 


original and unobvious applications. These decisions helped prompt 
recourse to a legislative remedy for an untenable situation (for those 
with a vested interest in the marketability of computer programs); in 
1980 Congress passed the Software Act granting some of the protection 
the judicial branch had been reluctant to offer but still leaving many 
issues unsettled. A Semiconductor Chip Protection Act followed in 
1984 with a new stii generis form of protection for chip masks (the 
templates from which chips are made). Neither copyright nor patent, 
this protection applies for ten years (less than copyright) and demands 
less originality of design than does patent law. In this case, the law itself 
replicates the ‘having -come from nowhere’ quality of the simulation. 
The Minnesota Law Review 70 (December 1985) is devoted to a sym- 
posium on this new form of legal protection for intellectual but also in- 
dustrial property. 

The Software Act began the erosion of a basic distinction between 
copyright and patent by suggesting that useful objects were eligible for 
copyright. In judicial cases such as Diamond v Diehr (1981), the court 
held that ‘when a claim containing a mathematical formula implements 
or applies that formula in a structure or process which, when considered 
as a whole, is performing a function which the patent laws were design- 
ed to protect (for example, transforming or reducing an article to a dif- 
ferent state of things), then the claim satisfies the requirements of [the 
copyright law].’ 

This finding ran against the grain of the long-standing White-Smith 
Music Publishing Co v Apollo Co decision of 1908 where the Supreme 
Court ruled that a player piano roll was ineligible for the copyright pro- 
tection accorded to the sheet music it duplicated. The roll was con- 
sidered part of a machine rather than the expression of an idea. The 
distinction was formulated according to the code of the visible: a 
copyrightable text must be visually perceptible to the human eye and 
must ‘give to every person seeing it the idea created by the original’*’. 

Copyright had the purpose of providing economic incentive to bring 
new ideas to the marketplace. Copyright does not protect ideas, pro- 
cesses, procedures, systems or methods, only a specific embodiment of 
such things. (A book on embroidery could receive copyright but the pro- 
cess of embroidery itself could not.) Similarly, copyright cannot protect 
useful objects or inventions. If an object has an intrinsically utilitarian 
function, it cannot receive copyright. Useful objects can be patented, if 
they are original enough, or protected by trade secrets acts. For exam- 
■ pie, a fabric design could receive copyright as a specific, concrete rendi- 

” This case’s relevance tion of form. It would be an ‘original work of authorship’ fixed in the 
for computer software tangible medium of cloth and the ‘author’ would have the right to 

litigation IS discussed *. . • , . , . . , - . 

in Peter Aufrichtig’s display It as an Ornamental or artistic object without fear of imitation. 

for°Smpmer™'^^^^^ fabric design, once embodied in a dress, can no longer be 

Programs in Read Copyrighted since it is now primarily a utilitarian object. Neither the 

Only Memory Chips’, (jpess, nor any part of it, can receive copyright. Others would be free to 
imitate its appearance since the basic goal (according to a somewhat non- 


fashion-conscious law) is to produce a utilitarian object meant to provide 
protection from the elements and a degree of privacy for the body inside 

What then of a video game? Is this an original work of authorship? Is it 
utilitariai; in essence? And if it is eligible for copyright, what element or 
aspect of it, exactly, is it that shall receive this copyright? The process of 
mechanical reproduction had assured that the copyright registration of 
one particular copy of a work would automatically insure protection for 
all its duplicates. Even traditional games like Monopoly, which might 
produce different outcomes at each playing, were identical to one 
another in their physical and visible parts. But the only visible part of a 
video game is its video display. The display is highly ephemeral and 
varies in detail with each play of the game. For a game like Pac-Man the 
notion of pursuit or pursuit through a maze would be too general. Like 
the notion of the western or the soap opera, it is too broad for copyright 
eligibility. Instead the key question is whether a general idea, like pur- 
suit, is given concrete, distinctive, expression. The working out of this 
distinction, though, lends insight into the degree of difference between 

‘Intelligent system’: the Macintosh II memory map. 


~ “ mechanical reproduction and cybernetic systems perceived by the US 

E N Kramsky, op cu, judicial System . 

^ ‘ For video games like Pac-Maii a copyright procedure has developed 

. that gives protection to the outward manifestation of the underlying 

214 US PQ 33, 7ih software programs. Registration of a copyright does not involve deposit- 

43 .’ ’P > > jng algorithms structuring the software of the ROM (read-only 

■ memory) chip in which it is stored. Instead, registration requires the 

deposit of a videotape of the game in the play mode.^° 

Referring to requirements that copyright is for ‘original works of 
authorship fixed in any tangible medium’, Federal District Courts have 
found that creativity directed to the end of presenting a video display 
constitutes recognisable authorship and ‘fixation’ occurs in the repeti- 
tion of specific aspects of the visual scenes from one playing of a game to 
the next. But fixing precisely what constitutes repetition when subtle 
variations are also in play is not a simple matter. For example, in Atari v 
North American Phillips Consumer Electronics Corp (1981), one District 
Court denied infringement of Atari’s Pac-Man by the defendant’s K. C. 
Munchkin. The decision rested on a series of particular differences bet- 
ween the games despite overall similarities. In elaboration, the court 
noted that the Munchkin character, unlike Pac-Man, ‘initially faces the 
viewer rather than showing a profile’. K.C. Munchkin moves in profile 
but when he stops, ‘he turns around to face the viewer with another 
smile’. Thus the central character is made to have a personality which 
the central character in Pac-Man does not have. K.C. Munchkin has 
munchers which are ‘spookier’ than the goblins in Pac-Man. Their legs 
are longer and move more dramatically, their eyes are vacant -all 
features absent from Pac-Man. 

This opinion, however, was overturned in Atari vs North American 
Phillips (1982). The Seventh Circuit Court found Pac-Man' s expressive 
distinctiveness to lie in the articulation of a particular kind of pursuit by 
means of ‘gobbler’ and ‘ghost-figures’, thereby granting broad protec- 
tion to the game by likening it to a film genre or sub-genre. The Circuit 
Court found the Munchkin’s actions of gobbling and disappearing to be 
‘blatantly similar’ and went on to cut through to the basic source of the 
game’s appeal, and marketability: 

Video-games, unlike an artist’s painting or even other audio visual works, 
appeal to an audience that is fairly undiscriminating insofar as their concern 
about more subtle differences in artistic expression. The main attraction of a 
game such as Pac-Man lies in the stitnulation provided by the intensity of 
the competition. A person who is entranced by the play of the game, ‘would be 
disposed to overlook’ many of the minor differences in detail and ‘regard 
their aesthetic appeal as the same’.^' 

In this decision, the Court stresses the process of absorption and feed- 
back sustained by an automated but intelligent system that can simulate 
the reality of pursuit. The decision represents quite a remarkable set of 


observations. The fetishisation of the image as object of desire 
transforms into a fetishisation of a process as object of desire. This 
throws as much emphasis on the mental state of the participant as on the 
exact visual qualities of the representation (‘A person who is entranced 
by the play of the game’). 

In these cases the courts have clearly recognised the need to guarantee 
the exclusive rights of authors and inventors (and of the corporations 
that employ them) to the fruits of their discoveries. Simultaneously, this 
recognition has served to legitimate the cybernetic metaphor and to 
renormalise the political-legal apparatus in relation to the question: 
Who shall have the right to control the cybernetic system of which we 
are a part? On the whole, the decisions have funnelled that control back 
to a discrete proprietor, making what is potentially disruptive once 
again consonant with the social formation it threatens to disrupt. 

Such decisions may require recasting the legal framework itself and its 
legitimising discourse. Paula Samuelson identifies the magnitude of the 
transformation at work quite tellingly, ‘It [is] necessary to reconcep- 
tualize copyright and patent in ways that would free the systems from 
the historical subjects to which they have been applied. It [is] necessary 
to rethink the legal forms, pare them down to a more essential base, and 
adjust their rules accordingly. It [is] necessary to reconceive the social 
bargain they now reflect. 

If eflbrts to gain proprietary control of computer chip masks, software 
and video games have prompted little radical challenge from the left, the 
same cannot be said for bacteria and babies, for, that is, the issues of pro- 
prietorship that are raised by new forms of artificial life and artificial 
procreation where the ‘social bargain’ woven into our discursive forma- 
tions undergoes massive transformation. 

The hidden agenda of mastery and control, the masculinist bias at 
work in video games, in Star Wars, in the reality of the simulation (of in- 
vasions, raids and wars), in the masculine need for autonomy and con- 
trol as it corresponds to the logic of a capitalist marketplace becomes 
dramatically obvious when we look at the artificial reproduction of 
human life. The human as a metaphorical, automated, but intelligent 
system becomes quite literal when the human organism is itself a pro- 
duct of planned engineering. 

Gametes, embryos, and foetuses become, like other forms of 
engineered intelligence that have gained legal status, babies-to-be, sub- 
ject now to the rules and procedures of commodity exchange. Human 
life, like Baby M herself, becomes in every sense a commodity to be con- 
tracted for, subject to the proprietary control of those who rent the 
uterus, or the test tube, where such entities undergo gestation. 

As one expert in the engineering of human prototypes put it, 
reproduction in the laboratory is willed, chosen, purposed and controll- 
ed and is, therefore, more human than coitus with all its vagaries and 
elements of chance.^’ Such engineering affirms the ‘contractor’s’ rights 
to ‘take positive steps to enhance the possibility that offspring will have 

Paula Samuelson, 
‘Creating a New Kind 
Property: Applying 
the Lessons of the 
Chip Law to 
Computer Programs’, 
Minnesota Lavt Review 
no 70, December 
1985, p 502. 

Cited in Christine 
Overall, ‘“Pluck a 
Fetus From its 
Womb”: A Critique 
of Current Attitudes 
Toward the 
University of Western 
Ontario Law Review 
vol 24 no 1, 1986, pp 

desired characteristics’ as well as the converse right to abort or terminate 
^"ibid, p7. offspring with undesired or undesirable characteristics.-' But what is 

most fundamentally at stake does not seem to be personal choice but 
power and economics. These opportunities shift reproduction from 
family life, private space and domestic relations to the realm of produc- 
tion itself by means of the medical expert, clinical space and commodity 
relations. The shift allows men who previously enjoyed the privilege of 
paying for their sexual pleasure without the fear of consequence the add- 
ed opportunity of paying for their hereditary preferences without the 
fear of sexual pleasure. 

Such ‘engineered foetuses’ and babies become so much like real 
human beings that their origin as commodities, bought and sold, may be 
readily obscured. They become t’ne perfect cyborg. As with other in- 
stances in which a metaphor becomes operative and extends across the 
face of a culture, we have to ask who benefits, and who suffers? We have 
to ask what is at stake and how might struggle and contestation occur? 
What tools are at our disposal and what conception of the human do we 
adhere to that can call into question the reification, the commodifica- 
tion, the patterns of mastery and control that the human as cyborg, the 
cyborg as human, the simulation of reality and the reality of the simula- 
tion make evident? 

Like the normalisation of the cybernetic metaphor as scientific 

Social power and cybernetic technology: an American advertisement for Micro-Systems Software. 

paradigm or the judicial legitimation of the private ownership of 
cybernetic systems (even when their substrate happens to be a living 
organism), the justification for hierarchical control of the cybernetic ap- 
paratus takes a rhetorical form because it is, in essence, an ideological 
argument. Dissent arises largely from those who appear destined to be 
controlled by the ‘liberating force’ of new cybernetic technologies. But 
in no arena will the technologies themselves be determining. In each in- 
stance of ideological contestation, what we discover is that the am- 
bivalences regarding cybernetic technology require resolution on more 
fundamental ground: that domain devoted to a social theory of power. 

Purpose, System, Power: Transformative Potential versus 
Conservative Practice 

Liberation from any literal referent beyond the simulation, like libera- 
tion from a cultural tradition bound to aura and ritual, brings the actual 
process of constructing meaning, and social reality, into sharper focus. 
This liberation also undercuts the Renaissance concept of the in- 
dividual. ‘Clear and distinct’ people may be a prerequisite for an in- 
dustrial economy based on the sale of labour power, but mutually 
dependent cyborgs may be a higher priority for a post-industrial post- 
modern economy. In an age of cybernetic systems, the very foundation 
of western culture and the very heart of its metaphysical tradition, the 
individual, with his or her inherent dilemmas of free will versus deter- 
minism, autonomy versus dependence and so on, may very well be dest- 
ined to stand as a vestigial trace of concepts and traditions which are no 
longer pertinent. 

The testing Benjamin found possible with mechanical reproduc- 
tion - the ability to take things apart and reassemble them, using, in 
film, montage, the ‘dynamite of the tenth of a second’ - extends 
yet further with cybernetic systems: what had been mere possibilities or 
probabilities manifest themselves in the simulation. The dynamite of 
nanoseconds explodes the limits of our own mental landscape. What 
falls open to apperception is not just the relativism of social order and 
how, through recombination, liberation from imposed order is possible, 
but also the set of systemic principles governing order itself, its 
dependence on messages-in-circuit, regulated at higher levels to con- 
form to predefined constraints. We discover how, by redefining those 
constraints, liberation from them is possible. Cybernetic systems and 
the cyborg as human metaphor refute a heritage that celebrates in- 
dividual free will and subjectivity. 

If there is liberating potential in this, it clearly is not in seeing 
ourselves as cogs in a machine or elements of a vast simulation, but 
rather in seeing ourselves as part of a larger whole that is self-regulating 
and capable of long-term survival. At present this larger whole remains 
dominated by parts that achieve hegemony. But the very apperception 
of the cybernetic connection, where system governs parts, where the 

Anthony Wildcn, 
‘Changing Frames of 
Order: Cybernetics 
and the Machina 
Mundi’, in Kathleen 
Woodward (ed), The 
Myths of Information, 
op cit, p 240. 

Gregory Bateson, 
‘Conscious Purpose 
and Nature’, in Steps 
to an Ecology of Mind 
op cit, p 437. 

Gregory Bateson, 
‘Style, Grace and 
Information in 
Primitive Art’, Steps 
to an Ecology of Mind, 
op cit, p 145. 

social collectivity of mind governs the autonomous ego of individualism, 
may also provide the adaptive concepts needed to decentre control and 
overturn hierarchy. 

Conscious purpose guides the invention and legitimation of 
cybernetic systems. For the most part, this purpose has.served the logic 
of capitalism, commodity exchange, control and hierarchy. Desire for 
short-term gain or immediate results gives priority to the criteria of 
predictability, reliability and quantifiability. Ironically, the survival of 
the system as a whole (the sum total of system phis environment on a 
global scale) takes a subordinate position to more immediate concerns. 
We remain largely unconscious of that total system that conscious pur- 
pose obscures. Our consciousness of something indicates the presence of 
a problem in need of solution, and cybernetic systems theory has mainly 
solved the problem of capitalist systems that exploit and deplete their 
human and natural environment, rather than conserving both 
themselves and their environment. 

Anthony Wilden makes a highly germane observation about the zero- 
sum game. Monopoly. The goal of the game is to win by controlling the 
relevant environment, the properties and the capital they generate. But 
Monopoly and its intensification of rational, conscious purpose masks a 
logic in the form of being ‘merely a game’ that is deadly when applied to 
the open eco-system. Wilden writes, ‘We usually fail to see that Mono- 
poly supports the ideology of competition by basing itself on a logical and 
ecological absurdity. It is assumed that the winning player, having con- 
sumed all the resources of all the opponents, can actually survive the end 
of the game. In fact this is impossible The Monopoly win- 

ner . . . [must] die because in the context of the resources provided by the 
game, the winner has consumed them all, leaving no environment (no 
other players) to feed on.’^’ 

‘There is the discovery,’ Gregory Bateson writes in one of his more 
apocalyptic essays, ‘that man is only a part of larger systems and that the 
part can never control the whole.’^* The cybernetic metaphor invites 
the testing of the purpose and logic of any given system against the goals 
of the larger eco-system where the unit of survival is the adaptive 
organism-in-relation-to-its-environment, not the monadic individual or 
any other part construing itself as autonomous or ‘whole’^’. ‘Transgres- 
sion does not negate an interdiction; it transcends and completes it.’ The 
transgressive and liberating potential which Bataille found in the viola- 
tion of taboos and prohibitions, and which Benjamin found in the poten- 
tial of mechanically reproduced works of art persists in yet another 
form. The cybernetic metaphor contains the germ of an enhanced future 
inside a prevailing model that substitutes part for whole, simulation for 
real, cyborg for human, conscious purpose for the decentred goal- 
seeking of the totality - system plus environment. The task is not to 
overthrow the prevailing cybernetic model but to transgress its pre- 
defined interdictions and limits, using the dynamite of the apperceptive 
powers it has itself brought into being. 

Policing Desire 

Pornography, AIDS, and the Media 


Where did you get the idea that AIDS is contagious, that it 
affects only intravenous drug users, gay men and blacks? 
Simon Watney argues that the western media have 
adversely influenced our understanding of AIDS and its 
effects. He also offers a provocative analysis of ways in 
which the image of the body is used in western society to 
determine our ideas of health, beauty and normality. 

Comedia November 164 pages Hardback 1851780 22 X £18.95 
Paperback 1 851780 238 £6.95 

Putting ‘Reality’ Tbgether 

BBC News 


How new is the news? Is news really the product of 
accidents of space and time, or is it rather the result of 
inescapable routines? Philip Schlesinger’s important 
new study explores how the ‘reality’ that we see on the 
BBC television news is constructed. 

December 336 pages Paperback 0 416 90190 5 £7.95 

Ttelevision Culture 


Television is unique in its ability to produce so much 
pleasure and so many meanings for a wide variety of 
people, lohn Fiske looks at television both as an agent of 
popular culture and a consumer commodity of cultural 
industries central to the practice of capitalism. 

December 350 pages Hardback 0 41692440 9 £20.00 
Paperback 0 416 92430 1 £7.95 


1 1 New Fetter Lane, London EC4P 4EE 




A Wisconsin project? The University’s emblematic badger brandishes the tools of his trade. 
(Collage courtesy of David Bordwell and Kristin Thompson.) 

* David Bordwell, Janet 
Staigerand Kristin 
Thompson, The 
Classical Hollywood 
Cinema: Film Style and 
Mode of Production to 
1960, London, 

writing a portion of a lengthy book like The Classical Hollywood 
Cinema I am of course flattered to find a reviewer spending so much 
efibrt and verbiage in discussing it. I could not help being surprised, 
however, to have my portion of that work suddenly taken up into a 
larger, apparently unified, and even completed entity called by Barry 

King the ‘Wisconsin project’^. Living and working in Wisconsin as I 
do, I realise that there is no such project, and I cannot help but think 
that some of King’s miscomprehensions and misrepresentations of the 
three books he deals with result from some notion that people who have 
studied or taught film at the University of Wisconsin-Madison have 
been in cahoots to perpetrate a monolithic ‘poetics of cinema’ on an un- 
suspecting world of film studies. 

To begin with, then, there is no ‘Wisconsin project’. Indeed, the 
amazing thing to me, in Madison for fourteen years now, is, given the 
large amount of work that has come out of this school, how little of it can 
be gathered together and seen as a unified enterprise. We have had pro- 
fessors and graduates publish on United Artists’ corporate history (Tino 
Balio), early animation (Donald Grafton), the Balaban and Katz theatre 
chain (Douglas Gomery), Alexander Dovzhenko (Vance Kepley), John 
Ford westerns (Peter Lehman), American avant-garde cinema (J J Mur- 
phy), Pop and Minimalist American film (James Peterson), theories of 
filmic closure (Richard Neupert), 1940s publicity (Mary Beth 
Haralovich), Nagisa Oshima (Maureen Turim), early film-iriaking in 
Colorado (Diane Waldman), and on and on, but little of it adds up to a 
whole. Probably the only people who can be said to have worked 
together consistently on a specific ‘project’ - an historical poetics of 
cinema and a history of certain aspects of dominant and alternative 
styles are David Bordwell and Kristin Thompson. Other faculty mem- 
bers and graduates have consistently shown considerable originality and 
independence in formulating research projects. Janet Staiger was asked 
by us to collaborate on The Classical Hollywood Cinema specifically 
because she had already demonstrated an expertise in the area of Holly- 
wood production practices, and that was a subject different from the 
ones on which we were working. Her participation in the project made it 
possible for the three of us to cover the classical Hollywood cinema more 
thoroughly - it did not simply reinforce some existing notion of that 
cinema. None of her subsequent work has been similar to either of ours, 
nor has it adhered to any ‘Wisconsin project’ - and we would hardly ex- 
pect that it would. 

Similarly, Edward Branigan has gone in distinctive directions which 
owe little to ‘neoformalism’ (though Branigan is certainly as well- 
informed about that approach as he is about many others). Indeed, 
though King may be surprised to hear it, we think it fortunate that there 
has been little or no institutionalisation of neoformalism, or the 
historical poetics of cinema, or ‘the Wisconsin project’, of whatever 
readers would care to call it, beyond our own work. We would like to 
think that, if there is a ‘Wisconsin’ anything, it consists of a group of 
scholars committed to exploring various historical and aesthetic aspects 
of cinema - but they do not, on the whole, work together in program- 
matic ways. 

It may seem that I am making too much of King’s phrase, ‘the 
Wisconsin project’. Yet the idea that three texts (written by four people) 

Routledgc and Kcgan 
Paul, 1985. 

^ Barry King, ‘The 
Classical Hollywood 
Cinema’, Screen 
1986,vol27 no 6, p 74. 
Further references to 
this article will be 
indicated in the text 
after toman numeral I. 
See also part 11, 'The 
Story Continues’, 
Screen Summer 1 987, 
vol 28 no 3, pp 56-82. 

^ David Bordwell, 
Narration in the Fiction 
Film, London, 
Methuen University 
Paperback, 1986 (first 
published by the 
University of 
Wisconsin Press, 


^ James Curran and 
Vincent Porter (eds), 
British Cinema History, 
London, W’eidenfeld 
andNicoIson, 1983. 

^ Martyn Auty and Nick 
Roddick, British 
Cinema Nont, London, 
British Film Institute, 

* Edward Branigan, 
Point of View in the 
Cinema: A Theory of 
Narration and 
Subjectivity in 
Classical Film, Berlin, 
Mouton, 1984. 

are unified forms the basis of his entire, lengthy review. Presumably, if 
the books are not parts of a unified project (and they are not, though 
Narration in the Fiction Film^ clearly holds many premises in common 
with The Classical Hollywood Cinema), then much of what he says is not 
valid. I will leave it to my colleagues’ individual responses to 
demonstrate how diverse their opinions can be despite large areas of 

Just as significantly, however. King begins and ends the first part of 
his review by implicitly contrasting ‘the Wisconsin project’ with ‘the 
material conditions of British academia’. American institutional sup- 
port, he states, made such a lengthy, heavily researched book as The 
Classical Hollywood Cinema possible. He contrasts it with such British 
books on British Film Year as British Cinema History* and British 
Cinema Now"^, opining that they are speculative rather than conclusive 
because no British institutions would pay for the kind of research that 
went into The Classical Hollywood Cinema. 

It is hard to tell whether this is simply disingenuous or deliberately 
slighting of British film scholarship. Of course, in some ways, it may be 
generally easier to do historical research in the USA because of the large, 
well-organised library system here; we also have more film archives 
(which are not infrequently used by British scholars). But much of what 
King says about British versus American institutions is inaccurate in 
relation to The Classical Hollywood Cinema. I am a self-employed writer 
and had no grant support while working on The Classical Hollywood 
Cinema. To travel for research and to write the book, I depended on my 
own income from writing and from a temporary half-time teaching posi- 
tion. I watched many of the relevant films in the very congenial facilities 
of the British Film Institute. Moreover, we could not find an American 
press willing to publish our manuscript; it was brought out by a British 
press, Routledge and Kegan Paul. Similarly, Branigan’s book was 
published by Mouton*. Thus two of the three books King is reviewing 
were not published by US presses, though he does not mention that fact. 

More importantly. King’s point about British versus American in- 
stitutions slights the work of many fine historians based in Britain. For 
all one could tell by reading his review, film history is virtually dead in 
the United Kingdom. What of Charles Barr, Richard Maltby, Robert 
Murphy, Vincent Porter, Sarah Street, David Welch, Nancy Wood and 
others? What about the government-supported project to add two 
volumes to the Rachel Low history of British cinema? (King may not 
ultimately like the results, but he can hardly deny that this project in- 
volves institutional support of a type at least equal to that available in 
the USA.) Moreover, until recently Britain had the only film journal 
devoted entirely to historical inquiry, the Historical Journal of Film, 
Radio and Television, based in Oxford. The various British Film In- 
stitute publishing projects have, alas, no equivalent in the States. (The 
American Film Institute offers virtually no support to academics.) 
Whatever differences we may have with Barry Salt on methodological 

and factual grounds^, there can be no denying that he has produced a 
major historical study, in Englandy and with little institutional 
support - a study which King refers to in passing but which neither he 
nor Screen has apparently found worthy of close analysis. (‘Exegetes’, we 
are told, would be rewarded by comparing our book with Salt’s - people, 
that is, who have some knowledge of the historical topics dealt with?) 

King’s paragraph (I, p 74) dismissing British institutions and scholar- 
ship, inaccurate and ungracious though it may be, bears an interesting 
relationship to the conclusion of his review: 

If, on the other hand, the authors are prepared to see the classical paradigm 
as a regulatory formation aimed at containing moments of ‘excess’ within 
Hollywood’s own practices, achieving only moments of unstable 
equilibrium, then most - but not all - of what is problematic in their account 
can be reconciled. Whether they are prepared to take this step is uncertain, 
but I suspect that in the future use of their work it will be taken for them. 
(I, P 88) 

^ See David Bordwell 
and Kristin 
Thompson, ‘Toward 
a Scientific Film 
History?’ Quarierly 
Review of Film 
Studies, Summer 
1985, vol 10 no 3, pp 

* See David Bordwell 
and Kristin 
Materialism and the 
Study of the Early 
American Cinema’, 
Wide Angle vol 5 no 3, 
1983, pp 4-15. 

In the past there has been a tendency for a certain type of Marxist film 
historian to seize upon previous historical work and to offer a new inter- 
pretation of the data presented there. Such writers basically rely upon 
data from older, more traditional historians, rewriting their accounts 
from a Marxist perspective. Bordwell and I have offered a critique of 
this tendency in relation to the early American cinema, as epitomised by 
Jean-Louis Comolli and Noel Burch. (Among other things, we suggest 
there that film history is at such an early stage that we cannot trust in the 
accuracy of most accounts so far published.®) Such historians have of- 
fered promissory notes for the establishment of a new view of film 
history, but we are still awaiting the result. 

Few of these rewriters of other people’s work have been so forthright 
about their activities as King. He calls for the old process to begin anew: 
Marxists should make ‘use’ of our work and rewrite it in a fashion that 
would be correct by his lights. King never suggests that these future 
historians (whom he drafts so blithely into work that he obviously is not 
keen to tackle) should go out and put in some time in the archives and 
libraries - just lift it all and straighten it out. 

Here is where King’s view of the lack of institutional support comes 
back. Presumably he thinks these future histories will have to take their 
data from books like The Classical Hollywood Cinema, because the 
British can’t get grants. Of course, our work was published in part so 
that future historians could use it as a source for their own work. But I 
cannot believe that King’s recommendations at the end of Part I add up 
to a healthy prescription for a vigorous Marxist rewriting of film 
history. (Whether we have at least contributed to such a rewriting is 
clearly a matter for dispute, but I cannot conclude that further work in 
this direction should follow King’s proposed lines.) 

These are my main general objections to King’s review of The 

^ Sec ‘The Formalist 
from Iowa, USA, 
Who Fell in Love 
with Ivan the 
Terrible’, Film NeTas, 
1985, p 10. 

Classical Hollywood Cinema. Most of his detailed analysis is devoted to 
the work of my colleagues, and I will leave it to them to answer him 
more specifically. Before closing, I have some responses to some local 
points King makes. 

The Classical Hollywood Cinema contains no discussion of ‘the evolu- 
tion of specific techniques’, as King states (I, p 85). Changes, yes; evolu- 
tion, no. Perhaps King is thinking again of Barry Salt. (Salt, surprising- 
ly, appears as a motif against us in King’s review. Little though King 
may sympathise with us, he should sympathise with Salt less - and I 
suspect this is one point upon which we and Salt would agree.) 

King concludes that ‘Thus it is necessary to assume that the criterion 
of aesthetic pertinence governs the scrutiny of the sample’ (I, p 86). This 
is not news, since it is one of the basic premises of The Classical 
Hollywood Cinema - as evidenced by a portion of the subtitle, ‘Film 
Style’. King need hardly treat as a point against us the fact that he has 
discovered one of our main stated aims. 

King quotes an interview with me’ stating that some of the deep 
focus in Citizen Kane involves shot/reverse shot and hence conforms to 
the Hollywood continuity editing system (I, p 87). He considers that this 
example is ‘striking, if not entirely consistent’ (I, p 86); ‘So much for the 
grain of the text’ (I, p 87). I presume he implies here that I am claiming 
in the interview that as a result of its use of shot/reverse shot, Kane is no 
different from other Hollywood films. Again, King is assuming that this 
mythical ‘Wisconsin project’ is all one thing. I would suggest that if one 
looks at the same film as part of two different studies and asks two 
significantly different questions about it, the fact that one comes up with 
different analyses of it does not constitute ‘inconsistency’. If one asks 
whether Kane violates Hollywood continuity rules, the fact that its deep 
focus generally uses shot/reverse shot would indicate that it probably re- 
mains within the wide range of possibilities allowed by the Hollywood 
system. (And if our book demonstrated nothing else, I hope it at least 
showed that the classical Hollywood cinema is not the monolith it was 
often held to be.) If, however, one is analysing Kane as an individual 
film, or an example of Welles’s work as an auteur, or whatever, one 
might wish to emphasise its differences from previous Hollywood films 
in its use of shot/reverse shot. After all, ‘the grain of the text’ does not 
exist outside history. It is relative and historically specific, existing only 
in relation to other films, genres, etc. The practice of placing individual 
films against a variety of backgrounds, when done with proper historical 
grounding, does not equal mere inconsistency. Moreover, to single out 
one element in a film or scene does not exhaust the film; it may contain 
many other elements, also relevant to ‘the grain of the text’. King seems 
to assume that a single scene can be used as only one type of evidence. 
Such a position would be empiricist in the extreme. 

We would never say, as King claims (I, p 82), that the fabula is ‘always 
given in advance of its technical embodiment’. The fabula is the product 
of the spectator’s active engagement with the syuzhet as embodied in the 

film as viewed. This mistake probably comes from King’s dependence 
on Fredric Jameson’s inaccurate account of the Russian Formalists’ 
work in The Prison-House of Language. *° 

King accuses us of adopting a position which ‘abandons’ meaning by 
subsuming it under the overall question of form (I, 82). We do abandon 
the idea of ‘content’, preferring to stick to the idea of ‘meaning’. But to 
‘subsume’ meaning into form hardly implies that we abandon or ignore 
it. Quite the contrary, all our analyses assume that meaning (including 
ideological meaning) is an important component of form. King again 
quotes a passage from an interview with me and manages to suggest that 
we are totally uninterested in meaning, although in that interview I em- 
phasise that neoformalism analyses meaning along with all the other 
components of a film. I am sure that we do not privilege meaning to the 
extent that King would wish, but this is no reason for him to claim that 
we ‘abandon’ it. 

King asserts that the criteria for successful technical innovation that I 
outline on pp 264 and 270 of The Classical Hollywood Cinema are ob- 
vious and hence, he claims, uninformative. ‘One would like to know 
what regime of visual representation would not require those qualities as 
basic' (I, p 83). There are at least three responses to this question. First, 
even if there is no such ‘regime’, that doesn’t mean that it is not useful to 
lay the criteria out explicitly. Most people probably do not think about 
them when considering the introduction of technologies. (And, given 
that King is calling for a Marxist rewriting of history, is it not one tradi- 
tional aim of Marxism to point out how very basic ideas have been 
naturalised in society?) Secondly, I do point out in The Classical 
Hollywood Cinema that during the era of the Motion Picture Patents 
Company the desire to acquire a camera that did not violate patents 
rights took precedence over considerations of visual quality and 
shooting efficiency. Thirdly, there is a multidimensional aspect to 
technical innovation. Two cameras, say, might have very similar char- 
acteristics, and the historians would have to look closely to determine 
why one was adopted and the other was not. Competing technologies 
have degrees of difference and tiny trade-offs in advantages. The Bell & 
Howell’s superior pilot-pin registration system competed against the 
Mitchell’s faster focusing system throughout the 1920s, and ultimately 
only the introduction of sound allowed the Mitchell to triumph, due to 
its quieter running. Such apparently minor advantages and disadvan- 
tages can only be compared if the historian has some notion of system- 
atic criteria for technological innovation. 

Apparently such specific historical points are not Mr King’s forte, as 
he leaves it to others to discuss whether our book actually gives a 
coherent, accurate and thorough account of Hollywood cinema. 

*®Fredric Jameson, The 
Prison-House of 
Language, Princeton, 
Princeton University 
Press, 1972. 



' Barry King, ‘The Story 
Continues: Barry King 
Returns to the 
Wisconsin Project’, 
Screen Summer 1987, 
vol 28 no 3, p 56. 
Hereafter, King’s 
review and our book 
will be referenced in 
this essay as follows: I, 
for King, ‘ “The 
Classical Hollywood 
Cinema”; A Review by 
Barry King’, Screen 
1986, vol 27 no 6, pp 
74-88; II, for King, 
‘The Story Continues’; 
and CllCyfot The 
Classical Hallyieood 
Cinema: Film Style and 
Mode of Production to 
I960 by David 
Bordwell, Janet Staiger 
and Kristin 
Thompson, London, 
Routledge and Kegan 
Paul, 1985. 

^ I hope that as a 
postmodernist King 
will appreciate my 
rewriting of one of his 
sentences (II, pp 56-7). 

GIVEN BARRY KING’S explicit allegiance to ‘deconstruc- 
tion, Lacanian feminism and postmodernism [sic] - [which challenge] 
the concept of a “master” narrative of rationality’ like a docile post- 
structuralist, I accept the constraint of considering this review on its 
own terms, the better, I think, to indicate what it systematically elides^. 
Hence, King’s reading of The Classical Hollywood Cinema must be 
praised for its contradictions, excesses, and lacks - in sum, its sympto- 
matic figures of desire. In fact, given my recent research interests in the 
historical spectator and the institutional, ideological, and psychoanaly- 
tical apparatuses constituting readings of texts, I would argue that this 
review says as much about the function of the reviewer-as-reader as 
about The Classical Hollywood Ci«ema-as-text. In that regard, an analy- 
sis of the King review is intriguing as an alternative to my own reading 
as well as other reviews of the book. Although my focus will be on Part I 
of the review, what I will do is implicitly and (now not so) covertly a 
gesture toward King’s notion of ‘a Wisconsin Project’ and to Part II of 
the review. 

Let me take, as an introduction. King’s central and admonishing con- 
cluding remark for Part I: 

This procedure, which makes style the 'Logos’ of an expressive totality, 
means that contradictory tensions within ‘classical’ film-making are sup- 
pressed and, given the global dominance of Hollywood, that alternative 
modes of representation are only beatable at the margin. If, on the other 
hand, the authors are prepared to see the classical paradigm as a regulatory 
formation aimed at containing moments of ‘excess’ within Hollywood’s own 
practices, achieving only moments of unstable equilibrium, then most - but 
not all -of what is problematic in their account can be reconciled. Whether 
they are prepared to take this step is uncertain, but I suspect that in the 
future use of their work it will be taken for them. (I, p 88) 

Now as another reader of The Classical Hollywood Cinema, I understood 
the text as not only having taken that step but providing ample evidence 

and argumentation as to how such a textual analysis could be related to 
extra-textual determinants. Perhaps Thomas Elsaesser’s synopsis of the 
book’s project in his review in American Film is close to my reading of it: 

What emerges is less how the industry saw itself or wanted to be seen by the 
public, than how it defettded, argued, and rationalized its working methods, 
behavior, and standard practices into a system of limits and possibilities, 
constraints and priorities, which were themselves expressed as rules - of 
what was correct and what wasn’t. On closer inspection, however, these rides 
turn out to be a necessarily and cunningly flexible game of give-and-take, 
whereby the industry both absorbs change and contains its disrupting effects. 
At the same time, Hollywood is able to reward individual excellence and 
talent, while protecting itself against its excesses. ’ 

^ Thomas Elsaesser, 
‘What makes Holly* 
wood KunVf American 
Fibut May 1985, p 53. 
(Italics in original.] 

‘ “The Classical 
Hollywood Cinema” 
Wide Angle 1985, vol 
7, no 3, p 76. 

Or as Tom Gunning reads the book for a Wide Angle review: 

The core of this work lies in this dialectical relation between technical and 
stylistic innovation and the enduring demand for narrative coherence. As 
stable as this model may appear, it is, in fact, a figure of regulated tension. 

. . . This regulation is a dynamic process, a process of change within limits. 

. . .A number of. . . technical or stylistic devices are investigated in the same 
way: first posing a problem for the classical paradigm, then undergoing a 
process of adaptation in which any potentially disruptive aspects are brought 
into line with narrative continuity. * 

Gunning’s latter remark also indicates that what is a textual process may 
likewise function diachronically through the era of Holl3rwood’s inter- 
national dominance. Thus, King’s search for a conclusion in the book 
that matches his expectations, bred out of Screen’s theoretical studies of 
the characteristics of the classical paradigm, is, for other readers, already 
met both at the site of the text and at a very specific historical level (that 
of periods located in groups of even a few years). 

Respecting the reader as a historically constituted subject, I would 
make no judgment as to the most valid interpretation, assuming as I do 
that multiple ideological, political, social, and psychoanalytical factors 
form the background for King’s and the others’ readings. Consequently, 
it would be presumptuous of me to argue that King’s reading of The 
Classical Hollywood Cinema is wrong (even if it seems at odds with the 
others or my own) since such a notion of ‘error’ disappears in post- 
structuralist ‘negative hermeneutics’. Finally, having no clinical or bio- 
graphical access to King-as-reader, my comments should be understood 
as addressed to the review-as-text, not to King-as-subject. 

Yet it does seem that some sketching out of the sites of difference 
between King’s and my reading of The Classical Hollywood Cinema 
would be beneficial for those other readers who may wish to consider 
any ‘future use’ of the work. In this regard, three areas of concern seem 
crucial for extended discussion: the problem of ‘primary determinants’ 

^ Taking a position on 
the text and the 
historical reader close 
to thatof.Tony 
Bennett in ‘Text and 
Social Process: the 
Case of James Bond’, 
Screen Education 
Winter/Spring 1982, 
no41,pp 3-14, 1 have 
difficulty with the 
notion of 'in a text’. 
For the purposes of 
this essay, however, I 
can point to the 
material evidence of 
sentences printed in 
the book, to be 
considered by ‘future 
users’ of the work. 

* At the most general 
level of history, I 
would argue that the 
primary determinant 
is economic - ‘in the 
last instance’ - but here 
we are dealing with an 
‘origin’ dated circa 
1895-96, and it seems 
to me that ‘the last 
instance’ is quite 
vague in terms of its 
effectivity (or 
explanatory power). 

in contemporary media studies; the notion of ‘class conflict’ in film his- 
tory; and the place of ‘difference’ in the classical Hollywood cinema. I 
shall suggest that where King’s reading and mine differ is in disagree- 
ments as to the logic of what was ‘in’’ the book and to what was - or was 
not - left out. As he puts it, he will ‘challenge’ the book ‘not on grounds 
of alternative evidence or an alternative reading of the same evidence, 
but in terms of the coherence of the arguments presented and their 
relationship to arguments not presented’ (I, p 75). I believe that his 
reading fails to do the former, although it might do the latter through 
challenging the book by projecting an alternative hypothetical history 
and through an alternative reading of the same sentences. Two related 
questions arise: how does King read the book? And, are other readings 
of it plausible? In a brief preview of the answer to the first question, 
King reads through strategies such as taking sentences out of context, 
ignoring companion sentences, conflating separate terms, and not 
seeing the conjunction ‘and’. 

I. ‘Primary determinants* in contemporary media studies 

One of the major thrusts of King’s reading of The Classical Hollywood 
Cinema is that despite the book’s apparent nod to economic determina- 
tion its interest in the impact that ‘style’ had on the mode of production 
ultimately produces a formalist analysis in which meaning is subsumed 
to form and ‘the “economic” becomes only the bearer of “style” ’ (I, 
p 81). Alternatively, I would read the book as privileging neither eco- 
nomic nor ideological/signifying practices as determinant. Three differ- 
ences exist between these readings. One is somewhat ambiguous. At this 
level of historical specificity’, I wish to allow for at least two ‘primary 
determinants’ while it is only clear that King considers ‘style’ fallacious 
as a primary one. In addition, I use the term ‘practices’ to underline the 
material activities in human praxis as opposed to any idealist ‘cause’. 
Most significantly, however, I employ the phrase ‘ideological/signifying’ 
as one of the two practices I believe are most pertinent to understanding 
the construction of the Hollywood mode of production. Surprisingly, to 
my knowledge, no reviewer of the book has made particular note of this 
distinction between my sections of the book and those authored by 
David Bordwell and Kristin Thompson. This phrase (used despite 
objections by my co-authors) is an excellent marker of the ‘incoherence’ 
of our text, but one King does not choose to examine since he seems to 
project our flaws elsewhere. 

Now to what extent, if any, my colleagues ‘unequivocally adopt’ ‘an 
approach which abandons the question of meaning altogether, usually 
by the mechanism of rendering all “content” an equal, hence redun- 
dant, expression of form’ (I, p 82), I know they are more than competent 
to answer. Yet I purposefully choose the phrase ‘ideological/signifying 
practices’ to emphasise 1) that what was being produced by Hollywood 

were cultural and discursive texts, with meaning-making as much a 
determining practice as the economic^; 2) that as a discursive act, films 
were extra-textually connected to, but not a one-to-one reflection of, 
their social formation; and 3) that ideology and signification should not 
be considered reducible to one another. It is, in fact, for me this practice 
which has been too often neglected in previous studies of the mode of 
production, to the detriment of historiographical advances in media his- 
tory. Earlier approaches to the American film industry have approp- 
riately noticed the industry’s relation to capitalist tactics of mass 
production but, as others before me have mentioned, distinctions must 
be made for cultural objects that operate overtly as signifying. Yet 
King’s reading of The Classical Hollywood Cinema ignores the explicit 
and potential implications of the use of this phrase in the sections signed 
‘Staiger’. If I have used this phrase, ‘ideological/signifying practices’ as 
one of two salient determinants for the mode of production and the 
industry’s products, then how can King suggest that my ‘account 
[which is] so apparently committed to the enumeration of concrete orga- 
nisational and institutional detail’ (I, p 77) becomes ‘a pretext for the 
demonstration of the primacy of style’ (I, p 77)? 

He argues this in two ways. First, he cites parts of a passage in which 
he quotes me as writing that ‘in the balance between economical produc- 
tion and a presumed effect on film, the latter won out (CHC, p 89)’ (I, p 
77). To what does ‘latter’ refer? It is King’s referent ‘filmic effect’, but 
the context of the quotation is concluding a specific example - the devel- 
opment of editing practices which included multiple workers, a 
sequence of labour, and written directions to maintain specific group 
norms of specific ideological/signifying practices materially and 
formally apparent through ‘style’. Symptomatically, the reading 
operates by displacement from the book’s discussion of editing to the 
reading’s cutting out of the sentence which follows the cited passage, the 
lack of which is telling for the reading’s representation of The Classical 
Hollywood Cinema. The sentence that follows is: ‘Thus, while economic 
practices helped produce a divided labor system of filmmaking, in many 
cases, ideological/ signifying practices influenced how the firms divided 
that labor’ (CHC, p 89). Not only does the deleted sentence operate to 
place economic practices in a primary position (a statement qualified by 
recognition of the function of adjacent media industries to serve as 
models for this industry’s mode of production) but it does not equate 
‘style’ with ‘ideological/signifying practices’. Figures of editing are, of 
course, ‘style’, but figures of editing are produced from historical and 
cultural notions of effective discursivity as well as functioning as 
bearers of traces of the ideologies (in the plural) which produced those 
notions. Hence I read this passage as suggesting not the ‘primacy of 
style’ but, if anything, a primacy of the economic practices and a 
potential and occasional tension (not a unification or symmetry) with 
ideological/signifying practices. Thus, this passage could be understood 
as rejecting any strong sense of economic determinism across both why 


^ Although I tend to 
appreciate Elsaesser’s 
comments that 
Hollywood might be 
better considered a 
service industry. Sec 
his ‘Film History and 
Visual Pleasure: 

Weimar Cinema’, in 
Patricia Mellencamp 
and Philip Rosen (eds). 

Cinema Uhtories, 

Cinema Practices^ 

Frederick, Maryland, 
Publications of 
America, 1984, p 49. 

and how labour was divided while remaining cognisant of its critical 
place in material history. 

In fact. King’s reading may seem to anticipate such an alternative 
understanding of the book. For later he writes, ‘I have suggested that the 
terms of the analysis of the “economic” in fact lead to the primacy of the 
“stylistic”, not by means of a denial of economic influence, which 
would be absurd, but by a relative negation of its full effectivity. The 
most obvious, as opposed to discursively embedded, example of this 
turns around the treatment of the concept of mass production’ (I, p 79). 
Yet, if one re-reads the above passage from The Classical Hollywood 
Cinema (p 89), one might perceive a telling distinction made between 
explaining why divided labour was employed (economic determination) 
versus how it was divided (economic and ideological/signifying). 
Actually, in re-reading my text of several years ago, I am somewhat 
embarrassed today to find this passage so economically determined, 
reconciling myself with the knowledge that overall I try to avoid a vulgar 
economic determination model. 

I would also recall here what the last twenty years of radical historio- 
graphers have attempted to avoid and that is earlier tendencies toward a 
grand synthesis, claiming one or a particular set of determinants as 
accounting for all historical process. Such monolithic theories of causa- 
tion or mechanistic structuralisms, as others have noted, operate in the 
vacuum of theoreticism. Such approaches ultimately impoverish our 
knowledge of the historical real. Hence, I read The Classical Hollywood 
Cinema as trying to specify how a capitalist economy and divided labour 
may have been signally involved with the overall choice of a mode of 
production. Yet the methods of dividing that labour cannot be linked 
just to capitalist exploitation of labour but also to ideological/signifying 
practices related to (but not totally or even uniformly) bourgeois systems 
of representation. On the contrary, I wonder how a solely economic 
determinationAjourgeois representation model of this medium’s history 
could explain how Hollywood’s labour was divided. But of course such a 
question is not necessarily implied by King’s reading since he takes no 
position regarding a preferred reading of the same or other evidence. 

The second way King reads The Classical Hollywood Cinema as sup- 
pressing the primacy of style is through my having written ‘a merely 
formalist treatment of the empirical details involved’ (I, p 77) since I 
choose ‘a material rather than a social definition of the concept of mode 
of production’ (I, p 78, italics in original). In reading this reading, it is 
not clear to me to what King’s attribution of ‘formalist’ in ‘formalist 
treatment’ applies. One way to take it is that the formalism arises from 
having chosen a particular type of definition (‘material’ rather than 
‘social’). However, he does not pursue how this choice results in ‘style’ 
becoming a primary determinant, shifting instead to queries about the 
place of class exploitation and the appropriation of surplus value in my 
theory. Whether or not I neglect these as he charges will be discussed 
below, but a material definition of the mode of production does not 

invariably - or even plausibly - result in the consequences of a theory 
that media history is determined by style. It can only occur in a reading 
which operates through condensing the notion of the type of definition 
(‘material’) with the projected historical determinant (‘style’). 

Another way to understand to what ‘formalist’ refers is to read it as 
describing the type of history being written; that is, my.method of writ- 
ing a history is ‘formalist’. Again, however, the question is how King 
can then read my primary determinant as style, and the answer is much 
the same as the above instance. This can be accomplished through con- 
flating the historiographical method with the meaning of what I am say- 
ing. In other words, my form (a formalist historiographical method - 
whatever that might be) determines my content (‘form/style’ determines 
film history). Perhaps we should welcome King to the ‘Wisconsin 

In thinking through these alternative readings of ‘primary determina- 
tions’ in The Classical Hollywood Cinema, it might do ‘future users’ well 
to observe several of King’s reading procedures (ignoring terminologi- 
cal differences among the various sections, quoting only parts of passa- 
ges, and conflating distinct notions) in operation in one other area of 
contemporary media studies - the question of determinants for a history 
of the conditions of reception. In King’s discussion of the book’s notion 
of what might be ‘an adequate history of the reception of the classical 
Hollywood film’ {CHC, p xiv), he writes the following paragraph: 

The normative ingredients of this mode of film practice also include certain 
assumptions about the activities of the spectator. But since the authors take 
the realms of style and production asprimary, these matters, plus the question 
of the concrete conditions of reception, are not accorded extetided treatment. 
For all that, the authors introduce certain issues cottcerning the spectator’s 
activity and the role of advertising in the establishment of classical canons, 
that will be ‘necessary’ in any future study of consumption (p xiv). The 
degree to which this is both an understatement and an overstatetnent will be 
explored in part two of this review. (I, p 76) 

Compare that with the referenced paragraph in The Classical Hollywood 
Cinema : 

If we have taken the realms of style and production as primary, it is not 
because we consider the concrete conditions of reception unimportant. Cer- 
tainly conditions of consumption form a part of any mode of film practice. 
An adequate history of the reception of the classical Hollywood film would 
have to examine the changing theater situation, the history of publicity, and 
the role of social class, aesthetic tradition, and ideology in cotistituting the 
audience. This history, as yet unwritten, would require another book, pos- 
sibly one as long as this. While we have not treated reception fully, the 
present book does introduce certain issues - e.g., the activities which the Hol- 
lywood film solicits from the spectator, or the importance of early advertis- 
ing in establishing classical canons - which we believe to be necessary to any 


future study of how the classical film has been consumed under specific 
circumstances. [CHC, p xiv) 

By deleting critical determinants including ‘the role of social class, aes- 
thetic tradition, and ideology’ (which would, in fact, suggest a represent- 
ation of history which is not a ‘formalism’), through recasting the thrust 
of the paragraph, and by failing to mark quoted phrases as taken from 
the book. King’s review appropriates the concepts introduced in The 
Classical Hollywood Cinetna as its own but different. As noted above, 
given the review’s allegiances to ‘deconstruction, Lacanian feminism, 
and postmodernism’, the review should not consider this an embarrass- 
ment although it is unclear how it fits with an expressed concern for ‘the 
forms in which surplus labour, surplus product or surplus value are 
extracted from the direct producers’ (I, 78). 

These strategies of reading are supplemented by another method 
which also impinges on possible interpretations of what constitutes pri- 
mary determinants in The Classical Hollywood Cinetna. King’s reading 
seems to work through constructing oppositions that may - or may not - 
be read as such. For instance, King writes: 

As a nominal concept, the mode of film practice is defined as consisting of 
two inter-related moments - a stylistic configuration demarcated by a ‘set of 
widely held’ -as opposed, one takes it, to hegemonically reproduced - norms 
‘about how a movie should behave, about what stories it properly tells and 
how it should tell them, about the range and functions of film technique’. 

(I, p 76, citing CHC, p xiv) 

First of all the original passage reads: 

A 7node of film practice, then, consists of a set of widely held stylistic 
norms sustained by and sustaining an integral mode of film production. 
These norms constitute a detertnittate set of assianptions about how a movie 
should behave, about what stories it properly tells and how it should tell 
them, about the range and functions of film technique, and about the activi- 
ties of the spectator. (CHC, p xiv) 

But more important is the inserted phrase, ‘as opposed, one take it, to 
hegemonically reproduced’. A reader could, of course, ‘take it’ as an old 
idealist Weltanschauung, but it is not clear to me why ‘one’ (unless she 
were King) must ‘take it’ as oppositional particularly since, as King has 
indicated (I, p 75), the book has signalled its allegiance to a Marxist cul- 
tural materialism and has just cited Raymond Williams as having ‘posed 
the problem’ of situating ‘textual processes in their most pertinent and 
proximate collective context’ (CHC, p xiv). In fact, ‘one’ might just as 
likely assume that the phrase ‘set of widely held stylistic norms’ and the 
‘determinate set of assumptions’ that they constitute might be referable 
to, say, Williams’ definition of ‘hegemony’. 

Now the theoretical model which I have beat trying to work with is this. I 
would say that in any society, in any paxiicular period, there is a central 
system of practices, meanings and values, which we can properly call domi- 
nant and effective In any case what I have in mind is the central effect- 

ive and dominant systems of meanings and values, which are not merely 
abstract bin which are organized and lived. That is why hegemony is not to 
be understood at the level of mere opinion or mere manipulation. It is a whole 
body of practices and expectations; our assignments of energy, our ordinary 
understanding of the nature of man and his [sic] world. It is a set of meanings 
and values which as they are experienced as practices appear as reciprocally 
confirming. * 

Now leaving aside the problems of King’s failure to define what he 
means by ‘hegempnically reproduced’ norms or any problems with 
Williams’ model, it is suggestive that The Classical Hollywood Cinema 
specifically notes how this determinate set of assumptions is repro- 
duced: ‘Moreover, this commitment to particular practices was not due 
to some essentialist “force” but rather to specifiable discourses discuss- 
ing, describing, and validating these practices’ (CHC, p 88). Not only is 
this model a reasonable translation of Williams’ concept of hegemony, 
but it is not at odds with a more generous reading of Althusser’s theoris- 
ing about Ideological State Apparatuses or notions such as Barthes’ 
work on mythology. It is, of course, different from the older, more vul- 
gar proposition of the infliction of the ruling class’s ideology upon the 
repressed class.’ 

Finally in regard to the issue of primary determinants, I would suggest 
that I very much like Elsaesser’s reading of this question: 

The authors’ approach tnight easily have turned into an exercise of mechani- 
cal Marxism and determinist historiography, to prove that Hollywood was 
simply the result of material facts - the capitalist exploitation of certain 
technologies - translating themselves into a bourgeois (‘realist’) form of 
representation that the need to return prof its on different kinds of investment 
shaped into standardization and rule-bound norms. Yet the overall thesis 
indicates, if anything, the reverse: both ‘style’ and ‘mode of production’ 
emerge as dependent on what I have so far deliberately kept off screen, so to 
speak, the fact that Hollywood is organized around the need and ability to 
tell stories. 

Consequently, ideological/signifying practices, channelled through and 
against story-telling (which stretches beyond the period of capitalism 
and across multiple modes of production), complicate the history of the 
American film industry as a simply determined medium in a capitalist 

* Raymond Williams, 
‘Base and 
Superstructure in 
Marxist Cultural 
Theory’, Nev> Left 
Reviez!) no 82, 
1973, p 8. 

’ Another instance of 
this tendency to 
construct oppositions 
that may be reading 
projections as much as 
anything else is in the 
review’s opening 
opposition postulating 
’what may be called the 
Wisconsin project’ as 'a 
strong foe’ to the 
‘material conditions of 
British academia’, and 
‘at a more immediate 
level’, to Screen (I, p 

Thomas Elsaesser, 
‘What Makes 
Hollywood Run?’, op 
cit, pp 54-55. 


II. ‘Class conflict’ in film history 

King argues that to ‘suppress’ the book’s real project of proving ‘the pri- 
macy of style’, my terminology and definitions (particularly for the 
mode of production) result in a representation of Hollywood which 
plays into this. In particular, he finds at least three problems. The first is 
that ‘the term mode of production is sometimes rendered as a rough 

equivalent to production practices ’ (Ij P 77) although at other times 

1 seem to follow ‘Marx’s specification’ by considering the mode of pro- 
duction as composed of ‘1) the labor force, 2) the means of production, 
and 3) the financing of production’ {CHC, p 89). However, secondly, 
King asserts that ‘Subsequently, any consideration of the form of 
finance,e.g. internal financing versus debt financing, or its source,e.g. 
banks or conglomerates, in the position of whole or part owner, is ruled 
out of consideration on the grounds, which can certainly be disputed, 
that this did not affect the management function’ (I, p 77-78). The con- 
sequences of these first two acts are to exclude from consideration ‘the 
provision of capital, whether in terms of plant or finance - not to men- 
tion the question of theatre ownership, distribution networks, etc ... ’ 
(I, 78). Finally according to King, I define the mode of production by 
using Marx’s 'material rather than a social definition of the concept of a 
mode of production. (The latter would emphasise the relationships of 
expropriation and the forms in which surplus labour, surplus product or 
surplus value are extracted from the direct producers.) Accordingly, 
those looking for a history of class exploitation in Hollywood should 
look elsewhere.’ (I, p 78, italics in original). King cites the works, ‘for 
example’, of Larry Ceplair and Steve Englund, The Inquisition in Holly- 
wood, and Michael Nielsen, ‘Toward a Workers’ History of the U.S. 
Film Industry’. 

Again, King’s account is an intriguing reading of The Classical Holly- 
wood Cinema. In this case, I might agree in part with him, for the book 
states: ‘ “Mode of production” will refer specifically to production prac- 
tices. The “mode” is distinct from the “industry” which is the econo- 
mic structure and conduct of the particular companies which produced, 
distributed, and exhibited the films’ {CHC, p 89). Here, in re-reading 
the text, I dislike the book’s phrasing for producing an unclear repre- 
sentation of the distinctions being laid out. Economists generally 
assume three structural levels for the American film industry: 
production, distribution, and exhibition. These may or may not be inte- 
grated formally or informally (through contractual agreements or 
merely a hand-shake). More significantly, though, each of these struc- 
tural levels has its own mode -or modes -of production. Obviously, 
structural relations among the three levels and among companies within 
the whole industry inform the management decisions at any specific 
point, but I would point out that it would be inappropriate (and sloppy) 
thinking to confuse the issues of industrial structure, conduct and per- 

formance with any specific level’s modes of production (labour force, 
means of production and financing). 

However, the rest of King’s assertions seem to operate from reading 
strategies similar to those employed for the question of ‘primary deter- 
minants’. Specifically, how can King’s reading produce the assertions 
that the book ‘rules out of consideration’ the forms of finance and their 
sources as well as class conflict? The ways include not distinguishing 
among concepts being discussed, ignoring the area of determination 
being examined, and projecting theoretical expectations. 

In the case of the question of finance, an alternative reading might 
focus on the difference between the notion of the forms and sources of 
finance and a particular type of economy - capitalism. Chapter 24 
entitled ‘The Labor-Force, financing and the mode of production’ 
begins with the assertion that ‘this chapter will look at two other aspects 
of the mode of production, the labor-force and the ownership and 
financing of the films’ (CHC, p 311). Having previously acknowledged 
as a trivial cliche that the film industry was within a capitalist economy 
{CHC, p 88), the book questions the implications of this fact for specific 
instances of decision-making at the management level of film produc- 
tion - but not for specific instances of decision-making at the manage- 
ment level for industrial structure, conduct, or performance in which 
case it might be pertinent. As the chapter continues, it does rule out of 
consideration as salient determining factors any specific forms and 
sources of capitalist financing because it extends the implications of capi- 
talist financing across a broader time frame. This results in capitalism 
having more profound effects than previously asserted but consequently 
less relevance as determining for any specific instance. In other words, 
the ‘logic’ of capitalism and the Holl}^ood concept of the quality film 
are so pervasive (or hegemonic) as motivating factors for production 
management - and labour - that it is meaningless to discuss any specific 
form or source of capitalist financing as a salient cause for management 
decisions, film practices, or production activities. Capitalism and this 
mode of film practice had so penetrated Hollywood from at least the 
mid-teens (and, I would argue, earlier) that it is superfluous to discuss 
any ‘deepening’ or variance of its effects. 

Is such a reading of the position of The Classical Hollywood Cinema 
possible? If this issue of financing is replaced within its context in 
chapter 24, it might be observed that it was being raised because I was 
disagreeing with earlier historians who imply that capitalist control is 
related to ‘the vertical integration of the major studios, the conversion to 
sound, and the unionization of the labor-force’ {CHC, p 31 1). The next 
sentence is ‘What traditional historians have tended to ignore is that 
labor-force activities and financing reinforced rather than contradicted or 
changed the production system as it had been constructed’ {CHC, p 31 1, 
italics in original). Now I believe that it is possible to read the text as 
excluding as pertinent the forms and sources of capitalist financing - but 

not financing or capitalism - as pertinent since the argument is that 
capitalism was relevant from the start of the cinema rather than appear- 
ing in the 1920s and early 1930s. Hence, no initially innocent cinema 
exists to be ‘captured’ by capitalists, either in the teens by the middle 
class or 1926-33 by a Rockefeller-Morgan sphere of influence. Capitalist 
profit maximisation goals operate from the start, inflecting management 
choices. Since King has excluded from his argumentation any provision 
for alternative evidence or alternative readings of the same evidence, I 
am left to ponder why he projects the theoretical expectation that films 
financed by banks would be significantly different from ones produced 
by the major studios. As I cited Robert Sklar in The Classical Hollywood 
Cinema : ‘there is “every indication that Wall Street’s interest coincided 
with that of Hollywood’s old hands -to make as much money as 
possible” ’ {CHC, p 315). I would underline that this position regarding 
the relevance (or lack thereof) of forms and sources of monies applies 
only to management decisions regarding the notions of what constitutes 
profitable films but does not apply to decisions affecting industrial con- 
duct and performance. (See C//C, pp 317-19.) 

Regarding labour-force activities, it is the case, of course, that labour 
was exploited and conflict occurred, but I would prefer a reading that 
distinguishes among the notions of the labour-force, real or social class, 
class identification, class conflict, and union activities, that recognises 
salient causes, and that thinks through the dialectics of theoretical 
expectations and historical findings. Recent radical histories of the 
working class in the United States have produced much material which 
does not correspond to earlier theoretical expectations. Since the 1960s, 
historians of American labour have tried to erase older equations of ‘the 
workers equal union members’ (a slippage due in part to the importance 
and power of the Commons School), to comprehend the wide extremes 
in political views and actions of the working class, to examine the dis- 
parity between real and imagined class identifications, and to determine 
the place of middle-line workers in the older ‘two-class’ version of his- 
tory. Yet, as I note in The Classical Hollywood Cinema^ ‘Outside the 
scope of my question (although an important problem) is the determina- 
tion of classes and the methods by which certain classes appropriate sur- 
plus value’ {CHC, p 89n), referring the reader to the theoretical work of 
Barry Hindness, Paul Q Hirst, Talal Asad, Harold Wolpe, and, later, 
Pat Walker, Nicos Poulantzas, and Erik Olin Wright {CHC, p 94n). 
Now why would 1 exclude the issue of classes from this question? Recall 
that what The Classical Hollywood Cinema is attempting to explain is the 
relationship between a very specific (and dominant) film practice and its 
historical context. It is not addressing as its object of study the history of 
labour in Hollywood. Consequently, what the book does need to con- 
sider is the pertinence of the fact of class and class conflict to this film 
practice, and it argues that the fact of class and class conflict is not signi- 
ficant to this historical question - not that it is not significant to history. 

Why is it not significant? Actually, the answer exists elsewhere in 
King’s review when he interjects the notion of ‘hegemonically repro- 
duced’ norms. Now if he uses that term in the older way, suggesting that 
a ruling class irnposes ideologies on the working class, then class does 
matter. King’s projected history of Hollywood would show that labour- 
ers in the working class held different norms of what the film practices 
should be and that union activities, other instances of class conflict, or 
hidden messages or aberrant techniques smuggled into the films were 
salient instances of the working class resisting the owners’ notions of 
what a quality film was. However, if he means by it the more current 
version something along the lines of the definition I used from Williams, 
then it is less apparent that real class is at stake although class identifi- 
cation might matter. In that case, what is to be studied is the labourers’ 
ideologies of ‘quality’ found through the material evidence of their dis- 
courses. And that was the function of citing manuals, articles and books 
by the Hollywood workers: indeed, if King wishes, a ‘para-ideology’ (I, 
p 87) that inhibits alternative practice. It goes back to the uncontested 
proposition that the members of the labour-force in Hollywood (no mat- 
ter what were their ‘real’ classes and despite the fact of classes and the 
appropriation of surplus value) shared very similar opinions about film 

One other disparity between King’s reading and mine is worth a 
special note: the effectivity of the unions in halting the incursion of new 
relations of labour and means of production that furthered fragmenta- 
tion and deskilling. In the history of the American film industry, con- 
flict with owners of the firms did appear most signally through unions, 
first through the lower work levels (projectionists, stagehands, carpen- 
ters and so on) and then into the middle levels (screenwriters, camera- 
men and actors). Unfortunately, the strategies that some specific unions 
employed only deepened their plight. The Classical Hollywood Cinema 
specifically discusses examples of competing unions scabbing on strik- 
ing unions and of negotiations among unions for strict jurisdictional 
boundaries which deepened the division among labourers (CHC, pp 
31 1-12). In fact, the two sources that King cites as alternative histories 
of Hollywood labour only provide further evidence of the failure of 
union activities to halt fragmentation or deskilling - or to arrest signifi- 
cantly the appropriation of surplus value. Ceplair and Englund on the 
HUAC period point out how politically conservative leaders of some 
unions could use red-baiting to crush radical organisations, and Niel- 
son’s essay (as well as his doctoral thesis) outlines a sad story of racketeer 
control of craft unions and payoffs to union leaders to keep labour in 

I am personally'^ appalled that this is what happened, but asserting 
that ‘craft unions, for example, do not reinforce the process of fragment- 
ation and deskilling - a grossly insulting formulation as anyone familiar 
with the history of labour struggles in the film industry will know - but 

" Also useful, 
particularly since it 
attempts a theoretical 
explanation of the 
facts, is Ida Jeter, 

‘The Collapse of the 
Federated Motion 
Picture Crafts: A Case 
Study of Class 
Collaboration in the 
Motion Picture 
Industry’, University 
Film Producers 
Association Journal 
1979, vol 31 no 2, pp 

Extremely personally, 
since as the daughter 
of a nearly 50-year 
member of the 
Union, I watched my 
father have to give up 
his craft (and his 
retirement funds) 
when his local union 
lost a strike 
attempting to halt the 
imposition of 
deskilling technology. 
As the first member of 
my family to even 
attend a university 
(much less to receive a 
PhD, linking me to 
the Wisconsin 
Project), I would 
rather that this piece 
of the history of 
unionism were not 
also part of my 

66 arrest its fullest implementation’ (I, p 80) cannot provide any coherence 

or scope to explain these events. Since King declines providing alterna- 
tive evidence, or even examining that supplied in the book, I suppose it 
would be outside the rules for our comparison of readings for me to 
request some explanation of the above assertion as it applies to the 
history of labour in the American film industry, based, as the assertion 
apparently is, on theoretical expectations about what ought to be the 
facts about craft unions. 

But on this topic. King’s line of reasoning in his reading can lead to 
another intriguing conclusion. He writes: 

If we allow that management can fail in this aim [of control over the labour 
process] or that sections of management are subject to the process of fragmen- 
tation as well, it still follows that somewhere in the work order agents are 
empowered to impart ‘meaning* to the flow of product across routinised pro- 
cedures. Such is the rationale forStaiger’s typology in the first place. But one 
needn’t accept the notion of ‘creativity’ or auteurisrn to recognise that 
which functionary, with what professional, social and personal character- 
istics, makes a difference to the way in which even a standardised ensemble of 
practices and techniques is activated. (I, p 79, italics in original) 

Since the point of his reading seems to be that the craft unions did pro- 
vide some kind of significant causality for Hollywood film practice, that 
this is constituted in ‘meaning’, and that ‘difference’ comes from spe- 
cific functionaries, then the ‘excess’ and ‘grain of the text’ he privileges 
might be as much from conservative and reactionary unionists as from 
any radicals. I suppose, however. King can read the difference. 

III. ‘Difference’ in the classical Hollywood cinema 

In fact, ‘creativity’ seems to be the subtext to King’s reading. For in his 
representation of the book, he focuses on only the first of the set of terms 
-‘standardisation and differentiation’ - not seeing the conjunction 
‘and’. My reading of The Classical Hollywood Cinema is that the book 
attempts to handle not only why Hollywood films exhibit a consistency 
at a more general level but also - and just as significantly - inconsisten- 
cies and differences are abundant at more discrete levels - incoheren- 
cies, inconsistencies and differences among genres, teams of workers, 
films, and within a film.*^ Standardisation, yes, but for this economic 
system, differentiation as well. If the weight of the argumentation and 
evidence seems to fall to the former, it is in part because at least this 
author of the book believed it to be the less well characterised in pre- 
vious histories and the more significant (theoretically and historically) to 
understand. The proposition of ‘sameness’ at one level of historical and 
theoretical discussion and the weight of argumentation does not, how- 
ever, necessarily need to result in a reading that the book does not 
explain difference or believes it is ‘trivial’. 

This seems to be 
King’s concept of the 
relative significance of 
‘sameness’ and 
‘ difference’ aswtU 
since his conclusion is 
that an acceptable 
version of ‘the 
classical paradigm’ 
would be one in which 
it is considered as ‘a 
regulatory formation 
aimed at containing 
moments of “excess” 
within Hollywood’s 
own practices’ (I, p 
88 ). 


Now, ironically. King’s reading not only wishes to believe that the 
book indicates that differences are non-pertinent but the reading’s alter- 
native history (which is being written in spite of - or because of - its 
denials*"*) attributes differences or excesses not to the system in interac- 
tion with human agents but, in a surprising move, ascribes it to 'which 
functionary, with what professional, social and personal characteristics’. 
Continuing the passage quoted above, the review states, ‘Naturally 
enough, it is possible to claim that such differences, not to mention less 
stable textual processes of “excess”, are trivial. But then the task is to 
show this by exploring the range of difference as well as sameness’ (I, 
P 79). 

Now where our readings of The Classical Hollywood Cinema might 
actually differ is not in the significance of needing to account for differ- 
ence but in that account (e.g., our explanations of it). King never spells 
out his position (he has excluded from his method of criticising us pre- 
senting ‘an alternate reading of the same evidence’). Consequently, the 
denial of resorting to theories of ‘creativity’ or auteurism but mention- 
ing human differences in ‘activating’ practices and techniques is all that 
is offered. This reading might seem the start of a humanism, an instru- 
mentalism, or a mechanistic structuralism. Moreover, the theory of 
‘activation’ is neither a dialectic materialism nor does it have as much 
coherency and scope as one suggesting an internal tension within the 
logic of capitalism: e.g., standardisation and differentiation. Quite 
frankly, I would still prefer explaining difference as due to human 
agency, consciously, non-consciously and unconsciously responding to 
(and against) a dialectical economic system (e.g., capitalism) that privi- 
leges consumption of the disposable (see CHC, chapter 9, ‘Standardiz- 
ation and Differentiation’, pp 96-112, and two of the book’s specific 
limit case studies - one, the subset of films labelled 'film noir'; the other, 
the specific film Citizen Kane, chapters 7 and 27). For me, difference is 
not trivial, but it is explicable, even if it is not predictable. 

The assertion by King’s reading that the book’s concern for ‘same- 
ness’ prevails can only be substantiated, as in the other cases, through 
reading strategies that are, as he would describe them, symptomatic. 
Take his extension of this point through focusing on the notion of a ‘fac- 
tory’ system in which he asserts, ‘And if Hollywood is like Fords - 
Staiger does argue that this is not literally the case, but the differences 
are not sufficient on her account to overturn the homology - then the 
same imperative “to insure the most efficient and economical work 
arrangement” can be taken to govern the mass production of cars and 
films’ (I, p 80, citing CHC, p 90). Now what differences does the book 
marshall to dispute the comparison? First, it introduces Marx’s distinc- 
tions between serial manufacturing (which characterised production of 
Hollywood films) and machine-tool (or modern) industry (which charac- 
terised production of cars). The book notes that the former did allow for 
‘some collective activity and cooperation’ among crafts people. The lat- 
ter has become the archetype of alienation in capitalism. Yet this 

*^ Having taken a 
position, the review 
should expect this. 

pointed claim as proof of difference between systems of production 
becomes in King’s reading ‘my “admission” ’ (I, p 80). 

In addition, the text emphasised product differentiation as critical in 
marketing and, inversely, on the mode of production. No matter how 
little one film might be different from many other films, Hollywood 
publicised that variation. Indeed, advertising is as much a part of this 
period of capitalism as is mass production. As I note, ‘Thus, difference 
and “improvement” in film practice was also necessary (for this reason 
filmmaking did not achieve the assembly-line uniformity prevalent in 
other industries)’ {CHC, p 109). I continue: ‘This tension [between 
standardization and product differentiation] results in two additional 
effects on the Hollywood style and production practices: the 
encouragement of the innovative worker and the cyclical innovation of 
styles and genres’ (CHC, p 109). That is, which ‘functionary’ is 
involved is not irrelevant, although ‘intentional agency’ is regulated by 
the industry’s norms. Furthermore, ‘excess’ should not be necessarily 
equated with class conflict but -just as plausibly - from a historical 
perspective, the worker accepting capitalism’s logic and ideology, and 
from a psychoanalytical perspective, the worker functioning in a 
particular psychic economy which may (or may not) reinforce 

If King’s reading blinds itself to the conjunction ‘and’, inverts claims 
to ‘admissions’, and projects onto the book a theoretical expectation 
which it is unwilling to amend, then it is possible to produce a reading 
which attributes to the text a reduction of significance to difference. As 
another instance of this, the review suggests: 

This development [of the continuity script] occurred against a background of 
growing demands for quality in production - specifically: narrative domi- 
nance and clarity, verisimilitude, continuity, stars and spectacle ([CHC], p 
96). Such demands, which at no point are considered in the light of their 
potential to clash or conflict (I, p 80) 

Such a conclusion can be reached if the reader skips passages regarding 
conflicts between norms of story coherence and causality and spectacle, 
stars and novelties {CHC, p 109), Bordwell’s extended treatment of this 
via bogus happy endings {CHC, chapter 7), and the discussion of the 
problem of ‘the balance between formula and showmanship’ {CHC, pp 

Two other examples wil round out this point. One is the review’s 
refusal to read ‘differentiation’ in the discussion of film technology. 
After praising ‘Bordwell and Thompson [for making] some substantial 
contributions to the history of film technology and practice’, it con- 
tinues: ‘it would be noted that the conception of standardisation set in 
train by Staiger’s contribution troubles these accounts at times’ (I, p 83). 
In my theoretical and historical discussion of technological change, I 
offer three reasons for the industry considering potential technological 

changes: 1) production efficiency, 2) product difTerentiation, and 3) 
adherence to standards of quality {CHC, pp 243-44), but I also suggest 
these reasons do not necessarily cohere and can at times be at odds. 

It is evident that these three causes for technological change can complement 
and collide with one another. Often, for instance, a change effected primarily 
for differetitiating a product [so much for privileging ‘sameness’] could be 
quite costly, particularly at the beginning. (CHC, p 245) 

In fact, this connects with King’s intriguing version of Thompson’s 
point about the criteria for camera technology. What he takes as ‘basic’ 
(I, p 83) for a regime of visual representation was not so naturalised at 
one point in the history of camera technology when cost was also consid- 
ered. King writes: ‘If the authors were concerned at this point with some 
epochal shift in the anthropology of visual representation, such observa- 
tions might have, if taken further, some explanatory bite’ (I, p 83). We 
should thank the reading for its help, but we thought we were interested 
in situating a specific social formation’s choice of tools and techniques 
within its given epoch of ideological/signifying practices. 

The other example is the review’s thesis that Bordwell’s analyses of 
the star and film noir fail as studies of (regulated) difference. This is 
because Bordwell’s ‘formulation seems to equate the potential for chal- 
lenge with the intention to challenge and to confuse the effectivity of a 
practice with its origins’ (I, p 85). This is an interesting criticism since 
two pages later King criticises Bordwell’s definition of ‘style’ from the 
inverse approach: ‘Ambiguously equating film with its plot and collater- 
ally plot with style allows the authors to treat meaning as an effect, 
rather than a constitutive moment, of the discursive practices of film- 
making’ (I, p 87). I suppose you could reconcile the disparities among 
these ideas if you assert, as the review seems to, that it is all right for 
‘meaning’ to be ‘constitutive’ but not ‘practices’. However, if ‘origins’ 
are not to be fallaciously confused with evaluating the effectivity of a 
practice (and they aren’t), then I am still left to wonder how to fit in 
King’s attribution of significance to ‘which functionary’. I don’t know 
that it is possible to sort out all of this, particularly since King’s review is 
not proposing an alternative history but examining the (in)coherency 
and gaps of The Classical Hollywood Cinema. Consequently, the ‘coher- 
ency’ and ‘completeness’ of his subtext does not matter. 

Actually, the question of ‘difference’ is not just its pertinence but also 
its place within the mode of production and the film practice. The eco- 
nomic system of capitalism and the practices of specific workers were to 
exploit innovations and ‘difference’ but also to recuperate that variation, 
playing with it, but then bringing it back into line within the dominant 
norms of story-telling or more specific generic conventions. The expla- 
nation for difference that I read the book as offering is an economic one 
although it may surprise King that I would also connect this to broad 
stretches of capitalist ideologies and to certain aspects of the uncon- 

scious. Yet such a general causation has little pertinence as an explana- 
tory model, and in The Classical Hollywood Cinema, I tried to provide 
theory and data at a more specific historical juncture - late nineteenth- 
century capitalism and, later, monopoly capitalism of the twentieth cen- 
tury in the United States in a mass-produced, discursive medium. 

In conclusion, I would recall that King’s ‘challenge’ was to be ‘Not on 
grounds of alternative evidence or an alternative reading of the same evi- 
dence, but in terms of the coherence of the arguments presented and 
their relationship to arguments not presented’. In reviewing the sites of 
difference between our readings of The Classical Hollywood Cinema, I 
would argue that King does not show any incoherence in the book’s 
argumentation. In fact, he finds it a very coherent project - albeit for- 
malist. However, he does seem to indicate some ‘arguments not pre- 
sented’. Here he disagrees with what he takes to be the book’s conclu- 
sions because the book does not adhere to his theoretical expectations 
regarding what might be relevant in what ways. Thus, the gaps he dis- 
covers for The Classical Hollywood Cinema are not technically ‘structur- 
ing absences’ as much as distances between his expectations for a history 
of Hollywood film and the book’s arguments. The causes for the 
review’s reading strategies may be bound up in these factors or they may 
be due to other reasons. At any rate, significant differences do exist 
between King’s reading and mine of The Classical Hollywood Cinema. 
Indeed, of course, it will be up to any ‘future users’ of the book to inter- 
pret it (and our readings of it) as suits them. 


Single Issue A$7 
(3 issues) 
Individuals: AS 1 6 
(overseas): AS18 
(airmail): AS30 
Institutions: AS40 
(airmail); AS48 

Current Issue, number 16, 1987 on the work of Agnes Heller, 

Please make 
cheques payable to 
Thesis Eleven 

Send to: 
Thesis Eleven, 
Dept, of Sociology 
and Politics, 
Phillip Institute 
of Technology, 
Bundoora, 3083 

The Human Condition 

Agnes Heller: Philosophy, Rational Utopia 

Agnes Heller 

and Praxis 

Freedom and Happiness 
Agnes Heller on rationality 
Agnes Heller and Our Needs 
Critical Theory of Communication 


Post-Marxist Radicalism 
Australian Radical Nationalism 
Australia and Sweden 

Richard Bernstein 

Peter Murphy 
Amadeo VigorelU 
David Cooper 

Sandor Radnoti 

Dick Howard 

John Murphy 

Francis Castles 

Plus reviews on Foucault’s History of Sexuality, the Whitlam 
Government and Stillwell’s Accord and Beyond. 

Previous Issue, number 15, 1986, includes: 

Social Democracy in Australia 

Stuart Macintyre 

Arendt’s Political Theory 

Ferenc Feher 

Praxis and Poesis 

Gyorgy Markus 

Foucault and Adorno 

Axel Honneth 

Plus Karl-Werner Brand pn new social movements, 
Michael Bittman on Marx and Weber and an 
interview with Cayatari Spivak 



' Note to non-US 
readers: WPA: the 
Work Projects 
Founded in 1935 by 
Franklin Roosevelt, 
mobilised unemployed 
workers in many Fields, 
ranging from building 
construction to the 
arts. It went out of 
existence in 1943. 

^ For ease of reference, I 
shall cite King’s essays 
as follows: *F refers to 
part one, “‘The 
Classical Hollywood 
Cinema”’, Screen 
1986, vol 27 no 6, pp 
74-88. ‘IP refers to part 
two, ‘The Story 
Continues . . . 

Screen, Summer 1987, 
vol 28 no 3, pp 56-82. 
Page references will be 
given in the text. 

^ Note to non-US 
readers: The badger is 
the state animal of 
Wisconsin. Every state 
in the US has by 
common custom its 
own animal, bird, 
flower, song, etc. For 
readers of Screen, the 
most pertinent analogy 
may be to the chapters 
of Ulysses. 

BARRY KING HAS revealed a hotbed of film research 
operating in the American Middle West. For the first time in public he 
has divulged our name - the Wisconsin Project (derived from a never- 
completed WPA dam*). More important, he has unmasked our junta' ^ 
relentless single-mindedness. ‘The fundamental thrust of the Wisconsin 
project is to establish how readers accomplish the identification of the 
cinematic image’ (II, 81).^ When we founded the Project, we swore an 
oath in badger’s blood ^ not to work on any intellectual problems but the 
one King has identified; and this oath was cast, believe it or not, in exact- 
ly his words. Admittedly, it is not clear what this sentence means. But 
whatever it means, it is our fundamental thrust. And we thought we had 
covered our tracks pretty well. Note how subtly we establish how 
readers accomplish the identification of the cinematic image etc, in, say, 
Staiger’s study of early screen-writing practices, or Thompson’s book- 
length arguments about how Hollywood film distribution gained inter- 
national hegemony, or my study of Japanese film history, or even 
Branigan’s account of sound in The Purple Plain.* The Project must 
now reckon with the awareness that all forthcoming work - Staiger’s 
historical study of reception, Thompson’s book of film analyses and her 
study of P G Wodehouse, Branigan’s further investigations into nar- 
rative, and my analyses of Ozu and of the conventions of film 
criticism - will be read as fundamentally thrusting in the manner 

Fortunately, however. King is far from fully disclosing our Fantomas- 
like operations. His investigation is marred throughout by howlers, 
misunderstandings, misrepresentations, obscurities and invalid in- 
ferences. In fact, incredible as it sounds, I can find nothing pertaining to 
my contributions that he gets right. In what follows I point out some 
major inadequacies of his account: his indifference to evidence (part 1), 
his misrepresentation of the texts and their theoretical underpinnings 
(2), his misunderstanding of how research-based arguments work (3), his 


faulty reasoning (4), his counterarguments (5), and his rhetorical mis- 
judgments (6). I conclude with some observations on the larger implica- 
tions of his enterprise, at least as they look from the Wisconsin 


On some key matters King reveals himself ill-equipped to judge the 
books under review. He displays no competence to evaluate any of the 
hundreds of empirical claims made in The Classical Hollywood Cinema'^ 
(hereafter CHC), Narration in the Fiction Filin^ (NiFF) or Point of View 
in the Cinema (POV). For all one can tell. King has never seen any films 
at all; or read any trade papers or technical journals, or indeed read very 
much theory outside the Screen approved list. (There are a few devia- 
tions, on which I shall comment at the end.) He mentions one film. 
Pickpocket:, but I cannot tell whether he has seen it, since he criticises my 
analysis by means of a quotation from, of all people, Robert Bresson. 

True, King has an inkling that the CHC resembles Barry Salt’s book 
on style and technology. He will not, however, compare the two works, 
since that is a mere exercise for ‘the exegete’ (I, p 75) ^ and King is no 
exegete, as you shall see. The value of such tasks, and indeed of first- 
hand research of any sort, is dismissed in inimitable Kingspeak: ‘It is to 
be hoped, and this is hardly a matter of doubt, that evidentially based 
assessments will follow in due course; indeed there is enough in CHC 
alone to occupy a platoon of theses’ (I, p 75). In other words: other peo- 
ple will do the dirty work; I’ll take the high road of Theory. 

Before we follow him on his climb, we ought to note that if you accept 
King at his word here, he cannot judge any of our evidence. How, then, 
can he assert that CHCTs history of film technology ‘makes some sub- 
stantial contributions’ (I, p 83)? His only authority on such matters, 
Barry Salt, would certainly not agree.® The real impulse behind this 
compliment is perhaps revealed in another sentence: ‘It seems likely 
that these findings . . . will withstand critique’ (I, p 83). This is only a 
bet - what Jonathan Ree describes as ‘watching the ways in which the 
currents of opinion are flowing, a kind of punting on what future auth- 
orities will say’’. 

King’s single passage through the land of data involves his discussion 
of the size of the ‘unbiased sample’ in CHC. He is distressed that the 
authors ‘did not attempt to balance the proportions of films coming 
from each of the Big Five and Little Three, let alone other sources of 
product’ (I, p 84). Of course, the point of our unbiased sample is exactly 
not to weight the selection by such a priori decisions. How a proper 
weighting might have been done King does not say. On what grounds 
would he determine how many films should represent Goldwyn or 
PRC? Despite the resources King claims to exist in the US, he will find 
it nearly impossible to see even obscure Paramount films, let alone a 

^ Janet Staiger, ‘Mass- 
Produced Photoplays; 
Economic and 
Signifying Practices in 
the First Years of 
Hollywood’, [Fide 
Angle vol 4 no 3, 1980, 
pp 12-27; Kristin 
Thompson, Exporting 
Entertainment: America 
in the World Film 
Market, London, 
British Film Institute, 
1985; David Bordwell, 
‘Our Dream-Cinema: 
Western Historio- 
graphy of the Japanese 
Film’, Film Reader 4, 
1979, pp 45-62; 

Edward Branigan, 
Point of Viev) in the 
Cinema: A Theory of 
Narration and 
Subjectivity in Classical 
Film, Berlin, Alouton, 
1984, pp 96-97. 

® David Bordwell, Janet 
Staiger, Kristin 
Thompson, The 
Classical Ilollyaood 
Cinema: Film Style 
and Mode of 
Production to I960, 
London, Routledge 
and Kegan Paul, 


® David Bordwell, 
Narration in the 
Fiction Film, London, 
Methuen, University 
Paperback, 1986. 

^ As insatiable exegetes, 
Thompson and I have 
discussed both 
theoretical and 
empirical aspects of 
Salt’s important book 
in ‘Toward a 
Scientific Film 
History?’, Quarterly 
Review of Film Studies 
vol 10 no 3, Summer 
1985, pp 224-237. 

® See our protracted 
exchange with Salt in 
Film Quarterly. ‘A 
Salt and Battery’, Film 
Quarterly vol 40 no 2, 
Winter 1986-87, pp 

59-62; ‘Reply to 
Bordwell and 
Thompson’, Film 
Quarterly vol 40 no 4, 
Summer 1987, pp 
59-61; ‘Salt ir,Fi7m 
Quarterly vol 40 no 4, 
Summer 1987, pp 

® Jonathan Rie, 
‘Marxist Modes’, Roy 
Edgley and Richard 
Osborne (eds). Radical 
Philosophy Reader, 
London, Verso, 1985, 

minor Republic film from the late 1940s. Moreover, it is worth men- 
tioning that despite the problem of film availability (which we discuss 
on pp 388-389), we have been more scrupulous and explicit about our 
selection procedure and data base than any previous researchers. King 
warns that ‘the claims for the unbiased sample should be treated with 
caution, not least because of the relatively small numbers involved’ (I, 
p 84). Proportionately, one would have to treat with utter scepticism 
those claims of other writers, from Kuntzel to Heath, who generalise 
about the classical system on the basis of a single film. Writers like 
Bellour, who relies on a few Hitchcock films and a couple of others, 
would gain only fractionally more credence. Yet I have not noticed King 
issuing a similar caution about the studios, periods, etc, not represented 
in these writers’ selection procedures. Even Barry Salt, who claims to be 
using a large and representative sample, has never published his data 
base, so the exegete whom King assigns to Salt’s book could not analyse 
his corpus as King has ours. He is able to doubt our results because we 
have spelled out the steps leading up to them; indeed, his very numbers 
derive from our breakdown (Cf/C, p 388). 

On the whole, however. King avoids empirical data - so assiduously 
that he declines to cite texts accurately. In two spots, he drops a ‘the’ 
from the original passage (II, pp 56 and 64). The missing signifier slides 
to another quotation and gets added to my text. For good measure. King 
changes all my later ‘a’s into ‘the’s (II, pp 59-60). A quotation said to be 
from page 22 of NiFF is actually on page 23, while one attributed to 
page 32 is on page 30 (II, p 59, and II, p 67). His citation ofMieke Bal’s 
Narratology seems to be inaccurate as well (II, p 68). And he attributes a 
quotation to page xii of CHC, but this page is blank (I, p 76). Is he here 
hinting at the emptiness yawning behind all signification? 

Taking literally the premise that the sign is arbitrary. King proceeds 
to revise my work. Where I had written that narrative is ‘the activity of 
selecting, arranging, and rendering story material in order to achieve 
specific time-bound effects on the perceiver’ {NiFF, p xi), he substitutes 
‘ordering’ for ‘rendering’ and ‘create’ for ‘achieve’ (II, p 57). Again, I 
wrote, ‘The same piece of information might have been conveyed many 
other ways, many of them requiring no sight or sound of Jeff at all’ 
{NiFF, p 49). King substitutes ‘in’ for the first ‘many’, and then 
eliminates the last clause without marking the ellision (II, p 69). (Since 
the original sentences are pretty awkward. I’d like to think he made the 
changes out of solicitude; but, considering his own handling of lang- 
uage, he is unlikely to have had improvement in mind.) Lopping off the 
tail end of sentences is one of King’s most common editorial devices: 
when he quotes our aims in the CHC (xiv), he leaves out of the list our 
reference to ‘the activities of the spectator’ (I, p 76). King’s painstaking 
attention to such matters may be gauged by the fact that of eleven in- 
dented quotations from Narration in the Fiction Film, he has altered the 
wording, punctuation, or attribution of four. True, he does not signal 

these difierences explicitly, but he knows how the Project likes to find 
such minutiae. We have assigned a platoon of theses to look for more. 



Carelessness and lack of familiarity with the data do not check King’s 
stride; he is bound for the high country of Theory, and scholarly com- 
petence is too heavy to take along. So let us consider how King fares in 
presenting ideas. 

He starts with the disadvantage of not knowing the theories at stake. 
Although he claims that Thompson’s and my arguments issue from 
Russian Formalism, King betrays no acquaintance with primary For- 
malist texts, relying instead on Tony Bennett’s New Accents crib and 
Fredric Jameson’s account. Both expositions were never particularly 
sound, and they are now outdated. From Jameson, King derives the no- 
tion that cinema is well-suited for the fabula/syuzhet distinction because 
‘film as a medium constitutes a material separation of form and content, 
the fabula being always given in advance of its technical embodiment’ (I, 
p 82). Whatever this means (is the fabula the profilmic event? the script? 
‘reality’?), it is wrong. Thompson and I explicity deny any such for- 
mulation. (More on fabula/syuzhet later.) King summarises our views 
with his usual fluency: 

A formalist position leads to a decisive emphasis on defining the specificity of 
a given medium as a means of signification', especially as a means of narra- 
tion, with techniques and their characteristic deployment being taken as the 
markers of the intrinsicality - the literariness or in Thompson’s neologism 
the ‘cinematicness’ ~ of the medium or media format. (I, p 82) 

In so far as this makes any sense, it is a complete distortion of our views. 
Formalism has no inherent bias toward signification (Formalism is not 
semiotics); it grants no special emphasis to narration; and techniques 
and their patterning do not necessarily mark the specificity of the 
medium (if that’s what ‘intrinsicality’ refers to). Anyone familiar with 
the writings of the Russian Formalists and the Prague Structuralists (I 
have to emphasise this, since King always ignores the extent to which 
Thompson and I rely on them) will recognise King’s capsule statement 
as gibberish. To take only one example: The Formalists are not especial- 
ly interested in defining a medium, say language; they are interested in 
describing a function, ‘literariness’, which is manifested in certain pro- 
perties and uses of language. Similarly, we are not interested in defining 
the medium of cinema. We are interested in analysing and explaining- 
historical conventions and functions. A description, analysis, or ex- 
planation, of narration or an 3 rthing else, is not a definition. (As we shall 
see. King’s critique depends extensively on confusing these concepts.) 

To forestall 
counterattacks, let me 
say that I do not 
subscribe to 
everything in 
Gardner’s excellent 
book; but then only 
dogmatists would 
expect me to. And lest 
King thinks that by 
citing an American 
cognitive theorist I 
pooh-pooh the British 
as backward, I happily 
refer to the important 
work of Richard 
Gregory, e.g.. Mind in 

Penguin, 1931, 
especially Chapters 
9-13 and P N 

Mental Models 
Cambridge University 
Press, 1983. See also 
Guy Claxton (ed). 
Cognitive Psychology: 
Nezo Directions, 
London, Routledge 
and Kegan Paul, 1980 
and Peter Lloyd et al. 
Introduction to 
Psychology. An 
Integrated Approach, 
London, Fontana, 

1 984, especially 
Chapters 4-6, 9-1 1. 

King even more disastrously misunderstands the other major 
theoretical influence on NiFF, contemporary cognitive theory. He bare- 
ly discusses this aspect of the book, apparently because he hasn’t the fog- 
giest idea what I’m talking about. He summarises it (and, inevitably, the 
Project in toto) as ‘information theory goes to the movies’ (II, p 82). 
Since Thompson’s discussion of camera technology and Staiger’s of the 
continuity script do not seem deeply indebted to information theory, a 
footnote adds defensively, ‘the comparison with information theory is 
not so far-fetched, given that Bordwell and Branigan, especially, use ter- 
minology and references within the field’ (II, p 27). The problem is that 
we don’t. If we were interested in information theory, you would find us 
discussing the ideas of Shannon and Weaver, Wiener, Cherry, Miller, 
Malmberg, Bateson, et al. Instead you find me referring to Anderson, 
Bruner, Fodor, Hochberg, Gregory, Neisser, Rock, et al. These are 
psychologists, not cyberneticians; they study perception and cognition, 
and if they refer to ‘information processing’, it is not at all in the sense 
that ‘communication engineers’ refer to ‘information’. Apparently at- 
taching footnotes and a bibliography to a book is of no avail, so I will 
simply mention that King will find a straightforward popular survey in 
Howard Gardner’s The Mind’s New Science: A History of the Cognitive 
Revolution (New York, Basic Books, 1987).*“ It reads as smoothly as a 
New Accents volume, but it differs in being much longer than 200 pages 
and in making extended, nuanced arguments. 

Undaunted by unfamiliarity with the pertinent theoretical frames of 
reference, King goes on to misrepresent the texts he reviews. Let’s start 
with his notion of my notion of narrative. He claims that CHC reduces 
narrative to ‘narrative logic’, or causality (I, p 88); but I insist that nar- 
rative logic includes parallelism as well as causality, and it needs to be 
represented in space and time too; all these are ‘narrative’. (See CHC, pp 
6-69.) He says I take narrative as simply a process of ‘selection’ (II, p 57), 
for which there is no warrant in my writing. In NiFFl do not say that a 
poetics of narration will ‘lead directly’ to an account of narrative activity 
in fictional cinema and then to a historical poetics (II, p 57); I say that 
the former ‘may encourage’ the latter {NiFF, p 336). The distinction 
among narrative representation, narrative structure and narration is not 
‘the metaphysical ground, in the Derridean sense’ of my study of narra- 
tion (II, p 57); it is only a set of enabling initial distinctions, entailing no 
metaphysical commitments. My point is borne out by another of his 
misrepresentations, that my analysis of narration ‘tends to equate narra- 
tion with the total order of the text minus destabilising moments of ex- 
cess’ (II, p 75). Had he been attentive to the first page oUNiFF, he would 
have noticed that narrative representation and narrative structure also 
contribute (non-metaphysically) to the ‘total order of the text’. King also 
claims that there is no possibility of change in CHOs tri-level model of a 
narrative film; but the model was devised exactly to explain change at all 
three levels; we explain how the model could do this (CHC, p 7), and the 
rest of the book seeks to prove it. (Maybe it fails to do so, but King 

doesn’t show that, since it would require some empirical investigation.) 
He says that CHC avoids discussing how form can ‘act as an alibi’ for 
specific social and ideological patterns; but pages 82-83 and 367-377 
discuss how classical narrative devices can be and have been related to 
just such patterns. 

Speaking of CHC, King persists in characterising if as proposing a 
model haunted by Hegelian idealism. It calls forth his most dazzling 
flightSj but compare what he says with what we say: 

King: The model has ‘a ubiquity and a capacity for transcendence’ (I, 


CHC: ‘The system cannot determine every minute detail of the work, 
but it isolates preferred practices and sets limits upon invention’ (p 

King: The model is ‘a reified absolute . . . logos of an expressive 
totality’ (I, p 88). 

C//C:‘My account here will construct the classical stylistic paradigm 
across several decades, emphasizing the continuity at the second and 
third levels. But by stressing continuity of function I do not imply 
that the systems’ paradigmatic range did not change somewhat’ (p 7). 
King: The model is ‘a seamless entity, devoid of internal contradic- 
tion’ (I, p 88). 

CHC: ‘In Hollywood cinema, a specific sort of narrative causality 
operates as the dominant, making temporal and spatial systems 
vehicles for it. These systems do not always rest quietly under the 
sway of narrative logic . . . ’ (p 12). ‘I have already suggested that 
compositional, generic, and realistic motivation do not always work 
in perfect unison, and I shall examine some typical dissonances in 
Chapter 7’ (p 21). ‘Narration can, however momentarily, break down 
the unity of the classical film’ (p 83). 

King: The model is ‘an ontological constant which exhaustively 
defines the studio-produced film’ (I, p 88). 

CHC: ‘No Hollywood film is the classical system; each is an 
“unstable equilibrium” of classical norms’ (p 5). 

The effect, he says, is to treat stylistic innovations as ‘minor recalibra- 
tions’ of a system that remains constant in its ‘formal relationships’ 
(I, p 87n); but approximately half the book is devoted to these innova- 
tions, which yield such minor changes as the soft style of silent cinema- 
tography, talking pictures, Technicolor and deep-focus cinemato- 
graphy. Each of these is discussed as not only performing canonised 
functions but also extending those ‘formal relationships’ that King 
claims we reify. 

The reader may have noted that many of the claims that King 
misreports occur quite early in the book, which does lead one to wonder 
how far he got into the text. The same musings strike me with respect to 
NiFF. He says I treat classical narration as ‘a tightly controlled move- 
ment from initial state to disturbance to restabilisation’ (II, p 66). But I 
don’t say that; Screen says that, and has been saying it for over a decade. 

Here is what I say, for instance, about the ending: ‘There are enough in- 
stances of unmotivated plot resolutions to suggest a second hypothesis: 
that the classical ending is not all that structurally decisive, being a more 
or less arbitrary readjustment of that world knocked awry in the 
previous eighty minutes’ {NiFF, p 1 59). I go on for almost a full page 
about ‘pseudoclosure’, which you might think would attract someone so 
interested in gaps, contradictions, fissures, seams, instabilities and other 
signs of damaged goods; but King has somehow missed it. (It’s in CHC 
too, on pp 82-83.) 

More blatantly, his characterisation of the narrational modes I outline 
(another half a book that has apparently gone undigested) bears virtually 
no relation to what I wrote. For example, he offers a ludicrously inac- 
curate summary of what I say about Soviet films of the 1920s and 1930s, 
claiming that they’re about ‘the dictatorship of the proletariat’ and that 
they ask the spectator to make ‘deductive’ connections (II, p 66)! His 
summary of the difference between ‘sparse’ and ‘replete’ parametric 
narration is hopelessly muddled. Asking how one can distinguish the 
two. King answers for me: ‘“Replete” narration tends to foreground 
stylistic events in a manner that “creates deviations in the film’” (II, p 
62). But if you look at the passage he is quoting, you find that I say: 
‘Establishing a distinctive intrinsic norm, either sparse or replete, may 
create deviations within the film’ {NiFF, p 285; italics for emphasis 

King also misrepresents my theory of spectatorial activity. He says 
that I characterise that activity as ‘hypothetico-deductive’ (II, p 62); I do 
not. In fact, it is plainly inductive, although it does use hypotheses. 
(Again, knowledge of the cognitive-theory literature would have saved 
King from a gaffe.) King says again and again that I offer a theory of 
‘reading’; but I explicitly say (p 30) that what I offer is not commensurate 
with the ‘reading’ model. (He disagrees, and I shall rebut him shortly; 
but he at least ought not to load me up with conceptual baggage I’ve 
deliberately jettisoned.) He says that in discussing parametric narration 
I counterpose ‘an “expert” reading to a “lay” reading in a manner 
which conceives of the latter commonsense or thematic reading as an er- 
ror’ (II, pp 74-75). On the contrary: the experts - critics like King - are 
the ones who cite Bresson as a clue to what a movie is about and who 
produce the thematic readings I argue against. 

On the same point, he says that I should offer a ‘realist’ reading ‘which 
explains the “surfaces” of the lay reading as a result of underlying tex- 
tual mechanisms’ (II, p 75). The audacity of this charge takes my breath 
away (briefly). The entire second half of NiFF aims to offer just such ex- 
planations, as in the chapter on art cinema (upon which King does not 
comment), where I suggest that ‘the art-film narration solicits not only 
denotative comprehension but connotative reading, a higher-level inter- 
pretation’ and go on to suggest how all the narrational processes I in- 
dicate invite interpretation (p 212). In a chapter he has purportedly read, 
I spend two pages showing how parametric narration in Pickpocket pro- 

duces just the sort of mystical ambiguity that King relishes, concluding: 
‘The sense of an order whose finest grain we can glimpse but not grasp 
helps produce the connotative effects of which thematic criticism 
records the trace’ (p 306). Had he read these passages, King might not 
have later grieved at my lack of ‘a realist epistemology’ (II, p 80). 

You can find lots more instances of misrepresentation. I am amazed 
that NiFF, so full of stills, diagrams, and minute descriptions of style 
(including an attempt to analyse a single shot of a Jancs6 film) can be 
described as a book which ‘underestimates, or more exactly analytically 
sets aside, questions of the materiality of the text’ (II, p 80).** Nor does 
the book ‘reduce ... the detective film to the dramaturgical premise 
“suspense”’ (II, p 70); if King will look at page 64, he will find a detailed 
account of the characteristic detective-tale syuzhet; he will also find that 
I emphasise the role of curiosity as much as that of suspense. Further, 
Branigan’s discussion of hypothesis/error accounts of narration does not 
‘match’ my discussion of enunciation and suture (II, pp 80-81); the com- 
parison is ‘clear enough’ to King, but to no one else: they are both criti- 
ques, but of different positions. King also claims that Thompson and I 
take an approach ‘which abandons the question of meaning altogether’ 
(I, pp 82); a glance at Film Art: An Introduction (pp 30-33 and elsewhere 
in the 1985 edition*^), or at any page of MFFwill refute this ludicrous 
assertion. Finally, in CHCv/e explicitly say (pp 10, 388) that we don’t 
have what King calls ‘a completely random selection procedure’ (p 86). 

Sometimes the Project just falls victim to his writing style. He says I 
say that ‘Film narration is composed of two distinct theoretical strands’ 
(II, p 58). No: I say that there are two theoretical traditions which seek to 
explain narration, which is a very different thing. He says I say that in 
parametric narration ‘Stylistic figuration must not rupture the hold of 
narrativity’ (II, p 72). Whatever that means, I am sure that it is wrong, 
and I know that I did not say it. And nowhere does anyone consider 
CZ/Cthe ‘culmination’ and ‘central text’ of the Wisconsin Project (I, p 
75); this is merely the puffery of a reviewer struggling to hold the 
reader’s attention. 

At other times. King distorts by omission. One could argue that he 
makes Ci/Ca more ‘theoretical’ book than it is by disregarding most of 
the historical material. In reviewing NiFF, King omits not only the 
discussion of the films (save Pickpocket) but most of the basic narrational 
properties I set out (knowledgeability, self-consciousness, communica- 
tiveness), the concepts of prominence and foregrounding, the argu- 
ments about temporal and spatial representation in cinema, and other 
significant matters. By ignoring my distinction between intrinsic and 
extrinsic norms. King can naively suggest that maybe what we need is 
an account of ‘a particular competence set up by the text itself’ (II, p 80). 
The concept of intrinsic norms aims to explain just such a skill, and if 
my individual analyses speak to any point, it is exactly to this one. Screen 
has surely changed if its reviewer has nothing to say about my discussion 
of Godard’s work, and I would have thought that my discussion of 

’ * King might reply that 
he has a loftier 
materiality in mind, 
that of ‘culturally 
given materiality’ (II, 
p 81), a fine phrase 
which is never 
afforded the luxury of 
a definition. 

David Bordwell and 
Kristin Thompson, 
Film An: An 
Imroductian, New 
York, Alfred A 
Knopf, 1985. 


perspective as a representational scheme would have elicited at least a 
mild yowl of protest. 


He worshipped general ideas and did so with pedantic aplomb. The general- 
ity was godly, the specific diabolical. - Nabokov, Pale Fire 

Perhaps King misrepresents these books in such detail because he 
misunderstands their, so to speak, fundamental thrust. He insists on 
grafting his ideas onto us; form/content, ‘reading’, the position of the 
subject. It does no good to argue explicitly against the adequacy of such 
notions, since he either misunderstands the arguments or just ignores 
them and continues to saddle us with views we don’t hold. At a deeper 
level, though, his eviction of the empirical has a lot to do with his inabili- 
ty to get inside our arguments. Since this is a point which touches on 
styles of theoretical argument more, generally, I shall expand on it a 

In relegating ‘evidentially based assessments’ to others, King thinks 
he is doing something rather simple: just looking at ‘the coherence of the 
arguments presented and their relationship to arguments not presented’ 
(I, p 75). Leaving aside his botching of the arguments, this assertion and 
King’s subsequent discussion assume that the ‘evidence’ is all one kind 
of thing. (Presumably it consists of those claims having descriptive pur- 
port or flagged with footnotes.) This allows King to flatten all claims to 
the same level of abstraction: the juggling of terms and concepts. But in 
historical or critical/analytical work, the evidence is ingredient to the 
texture of the arguments. Consider the levels of claims in an argument 
like this, which I schematise for the sake of demonstration: 

(a) Capitalism strives to control technological innovation. 

(b) As a capitalist industry, the Hollywood studio system sought to 
regulate technological innovation. 

(c) To ensure stylistic and economic stability, the classical system 
sought to innovate technology in a controlled fashion. 

(d) Such regulation was carried out through particular agencies, 
which used both discourses and practices to achieve their members’ 

(e) The Academy ofMotion Picture Arts and Sciences, the Society of 
Motion Picture Engineers, and the American Society of Cinema- 
tographers were crucial agencies in technological regulation. 

(f) Each agency had more specific or intrinsic economic/stylistic/ 
professional-ideological projects, and within and among these agen- 
cies incompatible strategies might emerge. 

(g) Such projects can be inferred from the discourses and practices of 
the agencies if we assume that the groups sought to regulate tech- 
nology and maximise their menjbers’ benefits. 


(h) Gregg Toland’s career bears the traces of the ASC’s contradic- 
tory injunctions to the cinematographer. 

(i) Citizen Kane can be seen as a problematic consequence of 
Toland’s professional situation and of the wider forces at play in 
Hollywood cinematography. 

(j) Citizen Kane was released in 1941, amidst a string of other ‘deep- 
focus’ films. 

(k) Citizen Kane's use of deep-focus was believed by many cinema- 
tographers to be excessive, but because of professional imperatives 
and Toland’s celebrity, a modified deep-focus style came into com- 
mon usage in the 1940s. 

Which steps in this argument are merely ‘evidentiary’ and which in- 
volve only ‘coherence’? There is not a single proposition without em- 
pirical import and not a single proposition without conceptual conse- 
quences. To assess this argument at all, you must weigh both inferential 
adequacy and empirical claims at every step. Of course in making such 
an argument, the writer runs great risk, since s/he may slip at any point; 
and in anatomising my argument here, I lay myself open to rebuttal on 
any of the propositions advanced. But that is what historical and critical/ 
analytical argument is all about. Since King never tries to get at the fine 
grain of an argument this way, it is no wonder that he has only this to say 
about the case in point: ‘See David Bordwell’s discussion of Gregg 
Toland and deep focus, p 345fT, which manages to attribute both great 
and little effect to Toland’s innovations’ (I, p 87n). CHC, NiFF, and 
POF are full of such arguments as this, and whether they’re adequate or 
not, one cannot assess them by simply hiving off something called 

King’s tactic is, however, characteristic of theoretical arguments 
nowadays. The trend is well described by Richard Johnson: 

The object is to show that a text is organised around a specif ic problematic. 
Certain problematics are held to be fundamentally flawed. If such a tendency 
is present - especially Althusser’s own trio of historicism, humanism or 
empiricism - it is held to exhaust the whole contetit of the text. The text falls. 
The procedure is a kind of intellectual lumber-jacking^ very exhilarating, 
especially when the target is some great big classical tree that has stood in the 
forest for many years. Down goes Weber! Down go the Marxist historians! 

Down goes Marx! But this mode of critique is almost wholly destructive 

and therefore non-accumulative: it problematises’ but rarely provides 
another substantive account, still less one that incorporates what was ra- 
tional in the first. It tends to'produce a search for ‘originality.’ It differs very 
much in these respects from more creative modes, of which Marx’s treatment 
of political econo7ny might stand as a model. 

All these consequences - the absence of a substantive alternative ac- 
count, the inability to grapple with the specifics of the position under 

^ Richard Johnson, 
‘Histories of Culture/ 
Theories ofldeology: 
Notes on an Impasse’, 
in Ideology and 
Cultural Production, 
Michele Barrett et al 
(eds), New York, St 
Martin’s 1979, p 55. 

consideration and the search for originality - are evident in King’s criti- 
que as well. 

King misunderstands our research initiatives in another way. For all 
his loyalty to Theory, he does not see that we pursue answers to ques- 
tions. To pose particular questions and subquestions not only focuses 
one’s energies; it commits one to as detailed and disciminating a set of 
answers as one’s questions and purposes demand. It also means not ask- 
ing other questions. Those questions may be just as important or in- 
teresting, but the researcher cannot answer all questions at once. Yet 
because one cannot, it is easy for a reviewer to raise lots of questions that 
a book does not ask, let alone answer. Consider King’s big complaint, so 
predictable that I could have paid him in advance to make it. What about 
ideology? he asks time and again. I frankly don’t know how to connect 
ideology, conceived in some non-vacuous way, with all the aspects of 
film history I have examined (at least not while achieving the density of 
texture which a historical or critical/analytical argument needs). If King 
thinks that every piece of research must answer questions about 
ideology, or subject positioning, or sexual difference (and what inciden- 
tally, would those questions be? Would everyone pose them in the same 
way?) then I guarantee that King will not like a lot of what I write: but I 
did not frame my questions in those terms. But King professes curiosity 
about such questions, so let him stir from the armchair. I invite King to 
show, with arguments at least as detailed as that sketched above, how any 
of my claims or evidence could answer a precisely posed question about 
ideology. He might warm up on this trifling problem: 

This trend of analysis probably accounts for what many may regard as the 
most extraordinary feature of the text, that its treatment of the role of 
ideology within the occupational confines of the film-making community 
(Hollywood) is confined to an examination of various ‘recipe’ texts, technical 
specifications and the like, that debate or prescribe how to deliver or render a 
narrative, an image, without exhibiting a great deal of curiosity about the 
social or political values of what is rendered (I, p 87). 

King’s astonishment that we don’t tackle this issue suggests that we real- 
ly missed a chance here. Assuming that the foregoing could be recast as a 
research question, I eagerly await King’s answer. 

King’s obliviousness to the problem-solving thrust of research shows 
up in another way. Many of the differences he claims to find between 
my contributions to CHC and my NiFF spring simply from the fact that 
the two books ask different questions. The former seeks to describe, 
analyse and explain how a particular group style emerged, functioned 
and changed. The concept of narration takes a place as one component 
in the overall answer. MFFasks what the process of narration in general 
entails and how one can construct a critically and historically infor- 
mative analysis of that process, especially in so far as it affects textual 
composition and spectatorial activity. These are related, but different, 

questions, and it is not surprising that the conceptual scheme shifts bet- 
ween the projects. King believes that you have one set of ideas you just 
apply to every problem that comes along. By contrast, one can see in- 
tellectual inquiry not only as involving change and self-criticism (King 
might note that I signal rethinkings of problems from my earlier work, a 
habit not conspicuously developed in the film theorists’whom he valor- 
ises), but also as involving a flexibility in recasting concepts to handle a 
problem at hand. King claims that an ‘extended’ definition of style ruled 
CHC and that a narrower (‘reductive’) conception is at work in NiFF. 
Now we shall see farther along that this is not the case. What I want to 
point out here is that because the concepts are used differently. King 
thinks that the treatment is ‘ambiguous’ (II, p 63). He can profess confu- 
sion at my confusion: ‘This is not even to debate the question of whether 
Classical Narration is a term that is conceptually congruent with 
•Classical Style, mode of film practice with narrational mode, and so on’ 
(II, p 64). I don’t know what ‘conceptually congruent’ means, but if it 
means ‘identical’, the answer is plainly no, and if it means ‘consistent’, 
the answer is plainly yes. The concepts have been constructed to answer 
different questions and should be considered in that light. 

By flattening out the arguments and ignoring the questions that in- 
form them, King tries to make the works under review seem like Grand 
Theory. For King, and perhaps others in the lumber-jacking trade, in- 
tellectual work consists of accepting a doctrine and then letting that 
generate a project. Whence the tendency to think that if the doctrine 
seems problematic, the conclusions can be dismissed. But for me at 
least, research begins with puzzles that I think worth solving, questions 
that others have not asked or have not answered to my satisfaction. This 
is not to say that I have no presuppositions or frame of reference. It is only to 
say that those presuppositions do not constitute a theory in any doctrinal 
sense. In asking questions, one can frame them so that they not only in- 
terrogate data but also reflexively test the adequacy of one’s presupposi- 
tions. (This is why, incidentally, Russian Formalism seems to me a good 
model: quite apart from its substantive insights, it refused to establish 
itself as Grand Theory.) King thinks that if one is out of sympathy with, 
say, the Slavic Formalist tradition, one will not accept our conclusions. 
This may be an accurate sociological observation, but it is not logical. If 
research is not a simple mapping of received concepts onto obliging 
evidence, then one should strive for an argument to the best explanation. 
You may disagree with Formalism as a ‘theory’, but if you recognise that 
the arguments address a significant problem, you, or somebody, needs 
to find a non-Formalist solution. For King in his tipster mode an argu- 
ment passes muster if it is likely to ‘withstand critique’; but since any 
argument can be criticised, this formulation cannot distinguish the 
grounds on which criticism can be withstood. The issue is whether an 
argument is stronger than currently available candidates. This is largely 
what ‘argument to the best explanation’ means. A powerful argument 

On Hollywood, see 
Lary May, Screening 
Out the Past: The 
Birth of Mass Culture 
and the Motion Picture 
Industry, New York, 
Oxford, 1980; Janet 
Wasko, Movies and 
Money: Financing the 
American Film 
Industry, Norwood, 
New Jersey, Ablex, 
1982; Richard 
Maltby, Harmless 
Hollywood and the 
Ideology of Consensus, 

. Metuchen, New 
Jersey, Scarecrow, 
1983; Robert Ray, A 
Certain Tendency of 
the Hollywood Cinema, 
1930-1980, Princeton, 
Princeton University 
Press, 1985. On 
narration, see Jean- 
Paul Simon, Le 
Filmique et le Comique, 
Paris, Albatros, 1979; 
'Enonciation et 

Communications no 
38, 1983; Roberto 
Campari, II Racconto 
del Film: Generi, 
Personaggi, Immagini, 
Rome, Laterza, 1983; 
‘Cindnarrable’, Hors 
cadre no 2, Spring 
1984; Michel Colin, 
Longue, Film, 

Discours: Prolegomenes 
a une Semiologie 
Generative du Film, 
Paris, Klincksieck, 
1985; Francesco 
Caseiti, Dentro lo 
Sguardo: II Film e il 
suo Spettatore, Milan: 
Bompiani, 1985; 
George M Wilson, 
Narration in Light: 
Studies in Cinematic 
Point of View, 
Baltimore, Johns 
Hopkins, 1986. 

challenges the reader to come up with another argument that answers 
the same questions at least as well. 

Of course one could argue that some questions are, on intrinsic or in- 
strumental grounds, more important or valuable than others. For exam- 
ple, King or the multitudes he claims to speak for (‘many may regard as 
the most extraordinary feature -...’) could simply refuse to consider any 
questions that don’t address ideology. Fair enough. If my questions 
mainly don’t, why waste time reading my work? It cannot just be the 
topics discussed. Many scholarly books on Hollywood or narration, 
some of them much more attentive to ideology, have been published in 
recent years, and King ignores them. (So has Screen}"^) So if we’re not 
asking the right questions, why pay any attention to our answers? 

I think the cause goes back, again, to the idea that empirical data can 
be extruded out of arguments. What the books under review produce, at 
one level, is a fairly large set of concepts and information concerning 
cinema. If one could only, using the correct theory and asking the impor- 
tant questions, incorporate this new material, one could have a better 
doctrine, or at least an up-to-date one. This is a recurring error in con- 
temporary theory, running back at least to Comolli’s belief that he could 
simply pluck out Sadoul’s and Mitry’s data and reinterpret them in a 
materialist light. To be consistent, the materialist researcher should 
hold that the theorist’s conceptual frame of reference at least filters what 
concepts and data are selected and perhaps even governs what concepts 
and data are recognisable.'^ 

King, leaving data drudgery to that platoon of theses, has no taste for 
working all this out. But, in my favourite passage, he does suggest the 
general conditions of the assimilation. If, he hints darkly, 

the authors are prepared to see the classical paradigm as a regulatory forma- 
tion aimed at containing moments of 'excess’ within Hollywood’s own prac- 
tices, achieving only moments of unstable equilibrium, then most - but not 
all - of what is problematic in their account can be reconciled (I, p 88). 

It is kind of him to allow us to pledge allegiance to his doctrine (even if 
he forgets that he lifts the ‘unstable equilibrium’ phrase from our own 
account (CHC, p 5). If only we had said this ourselves, and paid 
obeisance to the theory of the subject in a paragraph or two, we would 
‘withstand critique’, or at least almost. 

' ^ Take, as an example, 
King’s blithe 
reference to the ‘Big 
Five’ and the 'Little 
Three’ (I, p 94), as if 
these concepts did not 
emerge from work 
outside the Screen 
tradition, mostly that 
of Tino Balio and 


For someone purporting to work in theory. King commits some 
remarkable inferential misdemeanours. There are, first, the logical con- 
tradictions in his piece. King (inaccurately) claims, for instance, that 
NiFF holds that ‘all telling in film finally resolves into showing’ (II, p 
62). This attests to my eviction of a discourse-based account of narra- 


tion. Yet a few pages later, I am charged, along with co-conspirator 
Branigan, of harbouring a tendency ‘to read cinema from a linguistic 
perspective’ (II, p 81). 

Here is a more striking example. At the end of part 1, in the for-your- 
own-good warning quoted at the close of the last section, King seems to 
hold out hope that the Project might be taken into the Theory fold. But 
at the beginning of part 2, he is unequivocal: ‘Since my task is assess- 
ment, I can only gesture towards the radical incompatability of this pro- 
ject with these other discursive practices’ - i.e., ‘deconstruction, Laca- 
nian feminism, and postmodernism’ (II, p 56). Near the end of part 2, he 
finds a glimmer of hope in Branigan’s work: although Branigan may be 
‘complicit’ (more on this adjective later), nevertheless ‘those who wish 
to argue for the potential challenge of so-called non-representational 
signs (sound, colour, music) to a patriarchally structured subject, will 
find Branigan’s delineation of the limits of subjectivity in film useful’ 
(II, p 75). Now if we assume that the aforementioned deconstructionists, 
feminists and postmodernists are interested in challenges to a patriar- 
chally structured subject, then Branigan’s work is no longer radically in- 
compatible. Or perhaps ‘radically incompatible’ just means ‘different’. 
For as part II closes, we find the much meeker claim that ‘like it or not’ 
(I for one like it) the Project has produced ‘an alternative paradigm’ (II, 
p 82)- which hardly amounts to radical incompatibility. 

When King attempts conceptual analysis the results are yet more 
unedifying. For example, when I argue that the star and film noir pose 
no inherent challenge to textual stability. King produces this: ‘Such a 
formulation seems to equate the potential for challenge with the inten- 
tion to challenge and to confuse the effectivity of a practice with its 
origins’ (I, p 85). This is sheer bluster. The origins of the star system or 
the intentions behind Jibn noir are beside the point; I argue that the films 
themselves do not overturn the system. If this is not addressing ‘the 
potential for challenge’ and ‘effectivity’, what is? King asks of Thomp- 
son’s criteria for technological innovation ‘what regime of visual 
representation would not require these qualities [i.e., control, durability, 
portability, etc] as basic' (I, p 83). He does not understand that only by 
differentiating these criteria can we discriminate the very incom- 
patibilities and tensions that he thinks we don’t look for. (See Thomp- 
son’s reply on this point.) Moreover, ‘basic’ is equivocal, and is not 
clarified by being put in italics. Does King mean ‘necessary’? Obviously 
not, since many existing cameras have not offered much control, did not 
yield particularly clear or steady images, etc. Edison got along very well 
for a long time with a camera which was not portable. But now we have 
strayed from the heights of Theory down to the foothills of evidence, 
where King will not follow. 

King’s ignorance of the pertinent theoretical literature cooperates 
with a striking inability to pursue an inferential chain. For instance, he 
claims that I lack a realist epistemology (II, p 80). (Is this the first time a 
Screen writer has considered this a deficiency?) But it is a straw man. 

Douglas Gomery - 
who hail from . . . 

since the psychology which I propose is perfectly compatible with a 
realist epistemology. The Constructivist could argue that we construct 
inferences about the world in such a fashion that we gain knowledge 
about its underlying mechanisms. King tries to cast the charge in other 
terms: I have an excessively ‘conventionalist’ epistemology (II, p 70). 
But he is wrong again. A Constructivist psychology could argue that our 
construction of perceptual reality is not conventional, in the sense that 
all people in all societies automatically draw perceptual inferences in the 
same way. The difficulties here lie in King’s casual assumption that a 
psychology is reducible to an epistemology, and in his jumping to the 
conclusion that a Constructivist psychology entails solipsism, unique 
mental events, privacy of meaning, relativism, conventionalism, or 
whatnot. He is driven to invent a third term of contrast, one that he 
nowhere explains: my account ‘has failed to engage elements of an “en- 
vironmentalist” approach’ (II, p 81). Now the Wisconsin Project is 
proud of our environment (beautiful lakes, forests, hills and farmland - 
rather like Kent), but I fail to see its relevance to conceptions of spec- 
tatorial activity. 

I have already touched upon one of King’s most recurrent gaffes, that 
of confusing definition and description. He thinks that I define narrative 
poetics by the distinction among representation/ structure/ narration (II, 
pp 57-58), which is like saying that linguistics is defined by the distinc- 
tion among semantics, syntax, phonology, and pragmatics. He thinks 
that describing a mode of narration amounts to defining it (II, p 65). He 
thinks that I define parametric narration as a matter of camerawork when 
I use that technique (and others) to illustrate it (II, p 74). He takes a des- 
cription of replete and sparse options as a definition of parametric narra- 
tion (II, pp 74-75). 

Yet when I do offer a definition. King just ignores it. MPF disting- 
uishes between comprehending narrative films (constructing a fabula) 
and ‘reading’ them (ascribing abstract or symbolic meanings to them). 
King asks whether the real contrast is not ‘between different kinds of 
readings with different kinds of competences’ (II, p 68). This is like say- 
ing that baseballs and oranges are not really different things, but rather 
two kinds of baseballs. We draw definitional distinctions for particular 
purposes. In his desire to win a purely nominalist point. King ignores 
my purposes in making the conceptual contrast. Moreover, he does not 
scrutinise the contrast itself to see if it holds good, but he is so pleased 
with the term ‘reading’ that he uses it to characterise my, and others’, 
position throughout. (Recall that the Project’s fundamental thrust is ‘to 
establish how readers . . . ’ etc.) 

King’s problem with definitions is bound up with a larger confusion 
about how categories work in a theoretical argument. For instance, I 
claim that each mode of narration sets limits on what can be done within 
it. ‘It follows from this,’ King says with assurance, ‘that what is a com- 
positional option in one mode is ruled out in another’ (II, p 65). But this 
does not follow at all. My argument does not presume that categories are 

constrained by unique membership conditions, or by necessary and/or 
sufficient conditions. A goal-oriented protagonist is an option in Holly- 
wood films; so is it in Soviet montage films; so is it in parametric films. 
Similarly, King triumphantly claims that the analyses of the films con- 
tradict the narrational categories they are supposed to belong to. How 
dare I say that replete parametric narration uses syuzhet parallels when I 
have said that there are syuzhet parallels in a portion of Storm over Asia, 
which does not belong to the parametric mode? By now you can predict 
that King misses the point of my original claim. (The parallelism in 
parametric narration functions as redundancy to permit different 
stylistic options to stand out; the syuzhet parallelism in Storm over Asia 
holds the stylistic choices constant.) His basic error, however, is again to 
make the Carrollian'® assumption that no property can cross categories. 
A bat is not a robin, but both can fly, and the definition of ‘bat’ or ‘robin’, 
for certain purposes, might include mentioning aeronautical abilities. 

Indeed, for all his talk of defining. King is not much interested in dis- 
tinctions. Where I offer, at some length, specific accounts of different 
ways in which people make sense of films, he prefers to invoke some- 
thing called ‘competences’, which, needless to say, is never explained or 
defended as a superior conceptual construct. Where I offer detailed 
analyses of historically varying modes of comprehension, he would 
rather just say that some ‘readers’ are ‘smarter’ than others (II, p 82). 
The grand hills of Theory, supposedly the domain of nuanced distinc- 
tions and subtle analysis, turn out to be as flat as Illinois'^. 

So what does King think theoretical disputation is? Take one of his 
central points about C//C-that it makes the Hollywood system too 
stable. (That is a mild way of summarising King’s fulminations about 
‘expressive totalities’ quoted above.) Now our book offers arguments for 
why the Hollywood film and the mode of production are on the whole 
stable entities; we provide both theoretical propositions and empirical 
arguments to back this up. King offers no counter-propositions or alter- 
native empirical arguments - just the Grand Theory assumption that 
every film and every moment in film-making is teetering on the brink of 
self-annihilating contradiction. All he can do is repeatedly butt his 
presuppositions against our case. Such a tactic usually goes by the name 
of dogmatism. 

King also seems to believe that if I criticise a position I must help it 
find escape routes. In NiFFl mount a case against enunciation theory in 
cinema. King accepts my critique of the thinkers who promote enuncia- 
tion theory. But then blind faith takes over: ‘To establish that the search 
for enunciative markers is futile requires that other cinematic 
techniques - sound or music track, colour, or even casting -will not 
establish an inferential context in which unmarked “objective” shots 
are rendered “enunciative”’ (II, pp 60-61). Put aside the fact that the 
fundamental thrust of the Wisconsin Project is to establish the iden- 
tification of the cinematic image, and hence we wouldn’t dream of talk- 
ing about sound or music. Put aside the likelihood that if one cannot 

Note to US readers: I 
refer to Lewis, not 

^ Note to non-US 
readers: A state to the 
south of Wisconsin, 
not yet identified by 
Screen as harbouring a 

show any such markers in the camerwork or the profilmic event, it is 
unlikely that music, colour, or casting will prove very fecund along 
these lines. The simple fact remains that if one refutes what the most 
serious and accomplished thinkers have claimed about their theory and 
if one offers better answers to the same question, there is at present no 
good reason to believe the old theory. The burden of proof falls to the peo- 
ple who advocate it. If MacCabe, Bellour, Nash, Ropars, et al, have er- 
red, it is not my job to counter claims that they haven’t yet made. King 
clings to a theory which he acknowledges has, in the terms proposed to 
date, broken down. Let him fix it up. 

A detailed analysis of any passage of the essay would reveal a 
polyphony of conceptual elisions and muddles. As just one example, 
consider this typical passage: 

In the discussion of spectator schemata, prototype schemata which indicate 
generic variables are seen as less useful than template schemata which act as 
data processors (p 34). These latter processes, which are governed by a rather 
Kantian synthetic a priori ‘narrative structure’ over which sit the 
categories of space and time, are supratextual rather than transtextual. 
Consequently, one would look in vain for a substantive account of the 
‘reading’ process that engages with what Genette has termed a topic com- 
petence based on ‘the treasury of subjects and forms that constitute the com- 
mon wealth of tradition and culture’. Such a grounded competence, especial- 
ly pertinent to popular culture, barely engages the higher level ‘perceptual 
schemata’ which emerge as self-sustaining. (II, pp 70-71) 

The first sentence misrepresents the distinction between prototypes and 
templates; my examples of prototype schemata come from genre, but I 
also have used genre films to illustrate template schemata. Here is a 
typical case of King’s confusion between description and definition. 
And both sorts of schemata are ‘data processors’, so King’s opposition is 
conceptually groundless. Template schemata are not ‘governed by’ a 
‘narrative structure’; the latter is, again, an instance of the former. To 
claim that the narrative structure I describe is a Kantian synthetic a 
priori is a real howler in the light of such claims I make as these: ‘Several 
experiments yield evidence for the schematic function of a “template” 

of narrative structure in contemporary Western cultures I suggest 

only that the formats have heuristic value for analyzing narratives pro- 
duced and consumed in our culture’ {NiFF, p 35). And those ‘topic com- 
petences’ for which King yearns need further specification. At present 
they seem to comprise a list of everything we might know or believe 
which might be relevant to understanding a film; that is, a list of 
everything we might know or believe, period. (If King thinks that mere- 
ly studying Hollywood will strain the research resources of Great Bri- 
tain, what will this project require?) Anyway, once King supplies us 
with the set of such ‘competences’, the real issue is what can explain 
their acquisition and use. And then he will find himself once again fac- 
ing my claims about prototype, template and procedural schemata. If I 

am right, no such skills could be arrived at without schemata of the sort I 
describe. If this is not ‘direct engagement’ with the issue, what is? Once 
again. King has butted an unexamined assumption up against a 
developed argument. What he does not realise, finally, is that these 
perceptual schemata are not ‘self-sustaining’. Unlike film theorists, they 
require empirical data. 


To my charge that King does not grasp my arguments, he might reply 
that in three principal instances he counters claims set forth in CHCznd. 
NiFF. The issues at stake are: indexicality, the syuzhet/fabula distinc- 
tion, and the definition (again!) of discourse. 

King invokes indexicality in the hope of striking a blow against the 
‘conventionalism’ and anti-realism he imputes to a Constructivist ac- 
count. He does not, however, mention that NiFF explicitly rules the 
documentary film out of consideration (see p xiv)j I would argue that in- 
dexicality is undeniably an important factor in how people make sense 
of documentary films. The point is whether King shows that it is an im- 
portant factor in how people make sense of fiction films. He does not. All 
he does is quote Roland Barthes on the analogon, recycling a claim that 
Barthes had made fifteen years before and that Bazin had made decades 
before that (II, pp 69-78).'® A quotation from Barthes does not 
automatically carry the day. How do we know that the ontology, 
epistemology and psychology of the photograph carry over into the 
cinema? Lots of ink has been spilled on this issue, but King might start 
with V F Perkins’ Filtn as Film (a book which, upon its appearance in 
1972, was subjected, in these very pages, to lumberjacking techniques 
quite similar to King’s'^). Furthermore, there is no question that a 
Constructivist theory could explain indexicality (as a top-down applica- 
tion of a schema concerning how images are produced). But once more 
King does not attempt to confute my theory on these grounds - or, as he 
would say, ‘engage the issues’. He runs indexicality together with 
iconicity, says both apply to fictional cinema, and then, amazingly, con- 
nects both to my purported neglect of genre. (An entire chapter of NiFF 
is devoted to genre.) As it turns out, indexicality is just another handy 
concept to brandish threateningly. 

I despair of ever untangling King’s treatment of syuzhet and fabula. 
To put the matter briefly: the question to be answered is how spectators 
intersubjectively arrive at essentially the same ‘story’ when presented 
with a film. Both CMC and MFF argue that the process is one of con- 
structive inference. The particular construct arrived at is called the 
‘story’. Now terminology starts to diverge. In CHC, I argued that the 
text in toto offers us the ‘plot’, and, since not everything in the text can 
be said to prompt story-constructing inferences, the aspect of plot which 
cues the construction of the story I called narration {CHC, pp 12, 24). 

*® Roland Barthes, ‘The 
Message’, in Stephen 
Heath (ed and trans), 
Image-A tusic- Text, 
New York, Hill and 
Wang, 1977, pp 
17-18; originally 
published in 1961; 
‘Rhetoric of the 
Image’, in ibid, pp 
44-15; originally 
published in 1964; 
Andre Bazin, 
‘Ontology of the 
Photographic Image’, 
in Hugh Gray (ed and 
trans). What Is 
Cinema?, Berkeley, 
University of 
California Press, 

1967, pp 9-16; 
originally published 
in 1945. 

Sam Rohdie, ‘Review: 
“Movie Reader, Film as 
Film’* *, Screen Winter 
1972/73, vol 13 no 4, 
pp 135-145. 

Systematically organised aspects of the film medium, which I refer to in 
this context as style (more on this shortly), are said to be mobilised by 

After writing the book, I became convinced that this set of terms was 
conceptually clumsy (a fact I signalled in NiFF, p 344). First, because of 
the numerous terminological confusions about ‘plot’, it is better to use 
the term syuzhet. That called for the use of the corresponding term 
fabula', this I continue to treat as synonymous with the inferential con- 
struct, the story. Style is still considered to include aspects of the 
medium, and narration still cues construction of the story. The only con- 
ceptual difference, as I said in NiFF, is that ‘narration’ now occupies the 
place previously assigned to ‘plot’ and that ‘syuzhet’ now occupies the slot oc- 
cupied by ‘narration’. King makes heavy weather of this change, calling 
it ‘coy’ (I, p 87); but he does not understand it. Just to clarify it, here are 
two diagrams: 







Syuzhet Style 

King’s counter-arguments rest upon several misapprehensions. As far 
as CHC is concerned, he is convinced that I use an inconsistent concep- 
tion of style. But rather, I use two commonly accepted notions of style. 
First, there is ‘group style’ (p 3), which describes commonly used op- 
tions in storytelling mechanisms and the film medium across a body of 
films. Take German Expressionism as a group style: it involves not only 
certain recurrent features of setting, lighting, acting and so on, but also 
characteristic narrative devices and patterns of overall filmic construc- 
tion. When we discuss how an individual film utilises the resources of 
the film medium, we apply a second sense of ‘style’, that touched on 
above: the systematic use of film technique, which can exclude narrative 
devices and narrational patterns. The wide acceptance of this dual sense 
of ‘style’ may be gauged by Monroe Beardsley’s explanation: 

Monroe Beardsley, 
Aesthetics: Prohlems in 
the philosophy of 
Cnticism^ New York, 
Harcourt, Brace and 
World, 1958, p 173. 

When this term [‘style’] is applied to individual objects, it is best used to refer 
to recurrent features of texture [i.e., small-scale relations among 
parts ]. . . . When the term is applied to groups of objects, to the oeuvre of a 
given age or of a given painter, it usually refers to recurrent features of tex- 
ture and structure [i.e., large-scale relations among partsp^ 

In CHC, so as to obviate confusion, I used ‘style’ principally in its 
‘group-style’ sense and used ‘cinematic space and time’ to indicate those 
features of the film medium that might in other contexts (and without 

impropriety) be called ‘stylistic’. Mastering this double usage ought to 
be child’s play for any theorist who can, say, read Lacan or Foucault 
with understanding. 

In NiFF, the concept of style is applied in the second sense, that of the 
systematic use of techniques of the medium within the individual film. I 
indicate this explicitly: 

Style also constitutes a system in that it too mobilizes components - 
particular instantiations of film techniques - according to principles of 
organization. There are other uses of the term ‘style’ (e.g., to designate recur- 
rent features of structure or texture in a body of films, such as ‘neorealist 
style’), but in this context, ‘style’ simply names the film’s systematic use of 
cinematic devices. (NiFF, p 50). 

But King thinks this actually amounts to ‘circumlocutions for the ex- 
tended concept of style’ (II, p 63). Why? Because ‘if a use is 
“systematic”, it may pertain to a body of films and so on’ (II, p 7). True, 
it may under some description, but not under the one I stipulated; so 
where is the problem? King muddies the water still further by saying, 
incorrectly, that the distinction between group style and style of the in- 
dividual film corresponds to the distinction between extrinsic and in- 
trinsic norms. He also takes it that style now means ‘d la Barry Salt the 
enumeration of the use of devices’ (II, p 63). No: style is not an aggregate 
of devices, but a system. To top it all off. King recommends that the 
concept of style ought to be extended to include ‘set design, perfor- 
mance and so on’ (II, p 63). Who ever said not? In everything I have ever 
written, style has included such techniques. 

King’s capacity to generate phantom problems is even more evident 
when he discusses the fabula/syuzhet pairing. Although a diagram in 
MFF tries to show that the interaction of syuzhet and style prompts the 
construction of a fabula, and that this overall process constitutes narra- 
tion, King insists that the fabula is inferred from the syuzhet alone (II, p 
69). He thinks I should stick to using ‘story’ and ‘plot’ throughout, but 
on seeing how he confuses these (and misrepresents Meir Sternberg’s 
account, and tries to saddle me with E M Forster’s story/ plot distinc- 
tion), I decline his advice. 

King also insists that the fabula is onscreen. Here we re-encounter his 
difficulty with definitions. By stipulative definition, I make the fabula a 
mental construct. Therefore it is not on screen. But he thinks it is, or 
anyway, just a little: ‘Only if one defines the fabula as a mental event is it 
really meaningful to say that the fabula is never materially present on 
screen. In part it is, even though the syuzhet invests it with a “new” 
meaning’ (II, p 70). He suggests we redefine fabula as no longer a wholly 
mental construct. Then what is it? He does not provide a new definition, 
but he offers two reasons to think of it as material. 

1 . ‘Some of the signifiers on screen, if under the regime of representa- 
tion of the text, nevertheless have meaning in relation to conventions 

which are extracinematic, even if they present the “cinematic ap- 
paratus” with problems of working these conditions of reference 
through textually, e.g., location shots, physical shape of the seen en- 
vironment, actors’ physical characteristics’ (II, p 69). 

But none of this woolly passage counts against making the fabiila an 
inference or series of inferences; the cinematic/extracinematic distinc- 
tion is just irrelevant. I insist throughout NiFF that we apply ex- 
tracinematic knowledge to constructing the fabttla, but that does not 
make it material. 

2. According to King, the fact that ‘the cinematic figure is always to 
some extent an analogon' (II, p 70) and an iconic sign makes the fabula 
partly on the screen (II, p 70). 

But the image is in fact not always indexically bound to a referent; 
perhaps King’s disdain for empiricism has made him forget cel anima- 
tion, drawing and scratching and punching holes on film, etc. And if he 
says that these too are indexical analogous of the process of painting on 
film, drawing on film, poking holes in film, etc, then he will have to 
grant that painting and sculpture and printed language are analogous in 
the same sense (i.e., traces of the action of painting, sculpting and prin- 
ting); in which case Barthes’ claim about the uniqueness of the photo- 
graphic analogon falls. Similarly, the image is not always iconic, not 
even in ‘realist regimes’. (Has King seen Fantasia ?) Finally, it is embar- 
rassing to point out to a writer for Screen that a sign is not the object it 
refers to or denotes, and the material and form of the signifier are not 
identical with the referent. Whether what the image presents really ex- 
ists or not, there is no sense in which the fabula - a signified if ever there 
was one - can become a physical thing. 

Of course. King is free to reject my term, but nothing is clarified if he 
uses it to christen a sentient entity that is ‘in part’ physical and ‘in part’ 
mental. What this Frankenstein’s monster thinks about movies remains 
to be learned. 

One more topic elicits counterarguments from King, and they plunge 
into the same old problems of definition, equivocation and faulty in- 
ference. King does not like the way MFF treats the concept of discourse. 
I start from Benveniste’s definition of the concept in relation to the cor- 
respondent notion ofhisioire. I argue that both, especially the concept of 
discourse, have been unproductively and inconsistently broadened. 
This has led to conceptual confusions not acknowledged by the users of 
the term. King thinks that this treats discourse ‘in an overly empirical 
manner’. For discourse is actually ‘in the extended sense ... an in- 
terpellation process’ (II, p 60). Fine. Is it confined solely to verbal 
language? If so, how many other interpellation processes are there, and 
what are they? On the other hand, if interpellation includes all symbolic 
systems, why refer to an interpellation process rather than the interpella- 
tion process? And since so much hinges on definitions, how about fur- 
nishing one of ‘discourse’ in its extended sense - since calling it an in- 
terpellation process describes it but does not define it? Recasting big-D 

Discourse as ‘discursive formation’ a la Foucault is no help, since I will 
again ask how many such formations there are supposed to be, whether 
interpellation occurs via other processes as well, etc. 

I suppose that such questions reveal me to be ‘deceptively concrete 
and unnecessarily restrictive’ (II, p 60). So let’s get flagrantly abstract 
and unrestrictive. ‘There is a sense,’ King says, ‘in which all cinema is 
discursive, since cultural action implies a subject’ (II, p 60). In the same 
sense, all cinema is representational, since representation (as ‘cultural 
action’) implies a subject. In the same sense, all cinema is linguistic, in 
that language implies a subject. In the same sense, all cinema is sym- 
bolic, since a symbol implies a subject. In the same sense, all cinema is 
interpellative, since interpellation implies a subject. In the same sense, 
all cinema is an object, since all objects imply a subject. Does this make 
discourse, representation, language, symbol, interpellation and object 
identical concepts? I would be interested in whether King thinks there is 
any reason to keep these concepts distinct. If not, then we are back at 
category problems again, only reversing his earlier troubles; now if one 
concept shares any property with another, it becomes identical with it. 
Alternatively, if King thinks there is a reason to keep all these concepts 
distinct, then he ought to be receptive to my attempt to clarify their dif- 
ferences. In fact, though. King’s argumentative style answers my ques- 
tion. He prefers to repeat a litany of terms - representational regime, 
discourse, apparatus, competence, etc - without explaining them. Like 
idols carried around to make it rain, they are supposed to do work mere- 
ly by being invoked. 

In any event. King is not the best guide to the arid heights of Theory. 
His extended sense of discourse, no matter how generously interpreted, 
will not answer my criticisms ofMacCabe (who sees discourse as a set of 
signifying oppositions or as portions of a text structured vis-a-vis quota- 
tion marks) or of Belsey (who speaks of an ‘unwritten discourse’ in Bleak . 
House). Both writers think of discourse as a text-based phenomenon, but 
not in Benveniste’s sense. Finally, NiFF argues that Metz reverses 
Benveniste’s sense oi discours and enonciation in order to discuss cinema. 
King does not dispute my charge but tries to warrant Metz’s broader 
sense of ‘discourse’: 

Metz may be ‘metaphorical’ in his use of Benveniste but, in his tenns, the 
cinema is discursive (i.e., of the symbolic) prior to any specific act of enuncia- 
tion. This primary identification, as opposed to the secondary or ‘continued’ 
identification offered by narrative flow which constitutes BordwelTs level of 
enquiry is simply not engaged by the distinction between mimetic and diegetic 
modes of narration. (II, pp 59-60) 

Once more we have to clear away the misrepresentations of my case. 
Mimetic and diegetic refer not to modes of narration but to theories of 
narration. ‘Narrative flow’, if that means the activities within the 
diegetic world, is not my sole level of inquiry. And my distinction does 


Christian Metz, The 
Imaginary Signifier: 
Psychoanalysis and the 
Cinema, trans Celia 
Britton et al, 
Bloomington, Indiana 
University Press, 
1982, pp 49-50. 

capture Metz’s usage, for the straightforward reason that King misreads 
Metz. According to Metz, secondary cinematic identification is not with 
‘narrative flow’ but with the character in the fiction. Moreover, primary 
cinematic identification relies not on some vague quality of discur- 
siveness but on a set of relays. The spectator identifies with himself as an 
‘all-perceiving’ subject, and thus ‘as he identifies with himself as look, 
the spectator can do no other than identify with the camera, too, which 
has looked before him at what he is now looking at and whose stationing 

( = framing) determines the vanishing point His identification with 

the camera [is] that of a transcendental, not an empirical subject.’^' 
From such passages, and those I cite in NiFF, do I derive my claim that 
Metz’s diegetic or language-based conceptions of narration rely ultimat- 
ely on the mimetic assumption that narration requires an invisible, om- 
niscient camera. And this claim King has not refuted by his reference to 

Finally, word reaches me from Paris that in recent seminars Metz has 
been systematically criticising the application of the concept of enuncia- 
tion to the cinema. If this is the case, it will be exciting to see how 
theorists like King will square an Authority’s criticisms with their 
tenacious faith in an enunciation account of narration. 


By now several of King’s rhetorical tactics should have become evident: 
misstatement, oversimplification, equivocation, begging the question, 
etc. He will make a claim very boldly in the text and then hedge it in the 
footnote, or let loose a swarm of irrelevant or rhetorical questions (‘How 
adequate is the distinction between genre and mode? To what extent are 
narrational modes being depicted rather than modes of reading ...?’- 
II, p 72). He will then move on, implying that it’s impossible to iron out 
all the difficulties in my text. A tactic that merits special notice is his 
habit of pulling a remark out of context, playing with various ways of 
construing it, then considering another passage and expressing astonish- 
ment that some fancies derived from the first passage do not correspond 
to the second one. He can get a lot of mileage out of this, as when he 
claims that I make the telling/showing distinction ‘categorical’ 
(whatever that means) while Branigan renders it as two aspects of narra- 
tion (II, pp 79-80). King muses that perhaps I recast the telling/showing 
distinction as recounting/enactment in order to avoid an implied 
criticism from Branigan. But then he says that in fact it doesn’t really 
avoid such a criticism, since it ‘invites distinctions at the level of 
character action that the fabula/syuzhet contrast is ill-placed to address’ 
(II, p 80). What these distinctions are he does not say, but he hopes to 
leave the impression that he has gotten somewhere when all he has been 
doing is playing solitaire. 

One commanding and instructive aspect of King’s presentation is his 

Barry King, ‘Screen 
Acting: Reflections on 
the Day’, Screen May- 
August 1986, vol27 
nos3^, p 134. 

It is something of a rebuke, if a fortuitous one, that the diversionary excesses 
of British Film Year should have also seen the publication of the culminating 
tranche of a series of texts, more or less connected with the University of 
Wisconsin, that attempt, and in many ways succeed in, a major re-siting of 
the relationship between the poetics of film and cinema history. (I, p 74) 

Granted, King likes culminations: CHC is the culmination of the Pro- 
ject, just as a day school at the National Film Theatre was ‘the culmina- 
tion of a season of films on performance and film performances . . . 

But culminating tranche} Next we’ll be hearing about penultimate and 
anticlimactic tranches, and then where will we be? I would, however, 
like to hear more about the diversionary excesses of British Film Year. 
Who did what to whom? And did someone take photographs? Or at least 

The marks of decadence are so evident that I almost wind up believing 
in the evolutionary model of style. ‘Here syuzhet is no longer ... a 
category that engrosses narration* (II, p 63). (Narration has always been 
easy to distract.) Branigan’s approach ‘advances the theorisation of sub- 
jectivity which remains ingredient to the theory of ideology’ (II, p 75). 
(Is it the theorisation or the subjectivity which is ingredient? Or is it just 
me being unreasonably concrete and restrictive again?) ‘In relation to 
the specification of the spectator’s activity, Bordwell wishes to operate a 
distinction between a viewing and a reading’ (II, p 67). (Unlike others, I 
operate distinctions and make an automobile.) ‘Mr Style was in the for- 
tunate position of only finding in the cinema his own family* (II, p 63). 
(Read that over a few times before moving on to a kinship analysis of 
Comrade Fabula and M. Discours.) I did, however, feel like tugging my 
forelock at this compliment: ‘Bordwell is led again and again to fill out 
the analysis with (rather good as it happens) “commonsense” descrip- 
tions of character orientation and action’ (II, p 68). Whatever ‘character 
orientation’ means, encouragement like this can only cheer you along. 

use of language (or discourse, or the symbolic). A stylistician of art 
history might say that Screen went through a Renaissance (1970-1977), a 
Baroque period (1977-1982), and a Mannerist phase (1982-present)^ 
now, with King’s, essay, it passes into pure Rococo. Our Project oath, 
the fundamental-thrust sentence, with its ‘accomplish the identification 
of the cinematic image* phrase, is a fair example of his manner, but his 
essay’s opening salvo is still better. 


King’s essay, while marked throughout by idiosyncrasy, exhibits certain 
features characteristic of much theoretical writing in film and television 
studies. Let me indicate, notionally, some of those features: 

1 . Reliance on vaporous formulas rather than on explicitly con- 

structed concepts ,and propositions. King says, for instance, that I 
need ‘a’ theory of the subject. By this criterion, the research projects I 
admire most - Tynianov on verse, Shklovsky on prose, Gombrich on 
the visual arts, Genette and Sternberg on narrative, Baxandall on 
Renaissance painting - would fail. But what King really wants is a 
pledge of allegiance to one theory, that congeries of fuzzy, inconsistent 
and equivocal notions that allows him to produce something as gaseous 
as his review. 

2. Appeal to authority. The names Derrida, Barthes, Althusser, el cie, 
are dragged in to bless King’s critique. How he purrs when Branigan 
cites S/Z: ‘Obviously, given the tenor of my remarks as a whole, this is 
not an unwelcome connection’ (II, p 77). He does not consider the 
possibility that Barthes’ book is open to interpretations of which he 
would not approve; the mere mention suffices. Like many theorists of 
film. King dwells in the realm of scholasticism, where nothing can be 
said that is not ratified by reference to a sacred text. 

3. Inflation of terms and concepts. I have already supplied many in- 
stances, but here is a staggering one: Branigan is chided for not being 
able to address ‘cinematic representation in toto - anthropomorphic as 
well as zoomorphic and cosmomorphic representation’ (II, p 75). Here 
Discourse Theory anticipates the Second Coming. 

4. Intellectual provincialism. The ignorance of Formalism and 
cognitive theory is one example. Another is King’s claim that NiFF is 
not commensurate with ‘psychoanal 3 ?tic accounts of cinema’ (II, p 71); 
but he plainly has in mind only Screen’s versions of Lacan. There are 
various sorts of psychoanalysis - something to be borne in mind by the 
theorist who automatically begins sentences with ‘Psychoanalysis 
teaches us that . . . ’. 

5. Critique as rectification. I don’t mean only rectification in its 
Stalinist form, as when King weighs Branigan’s argument for its ‘com- 
plicity’ (II, p 75). Rectification has a more opportunistic side, when the 
‘lumberjacking’ technique turns into a salvage operation: translate any 
intriguing empirical findings into the terms of the preferred dogma, 
then reject disagreeable findings on grounds that they don’t square with 
it. Since classical texts must be ready to break down, any account that 
argues that they aren’t so fragile gets rejected summarily: ‘Whether [the 
authors of CHC\ are prepared to take this step is uncertain, but I suspect 
that in the future use of their work it will be taken for them’ (I, p 88). 
We supply grist for the Theory mill to grind out the same old Contradic- 
tory Text. 

Nevertheless, King’s essay betrays signs of change. For one thing, his 
mention of Derrida, his invocation of analog/digital jargon, his Bazin- 
flavoured view of referentiality, his appeals to realism and anti- 
conventionalism, and his savouring of ‘the emergent richness of mean- 
ing’ as testified to by Bresson’s own words - all this suggests that differ- 
ent, and one might say ‘contradictory’ winds are blowing in the upper 
reaches of Theory. Moreover, the terrain has shifted. Now subject- 

position theory must compete with powerful splinter groups. There are 
the Postmodernists, who lack the proper devotion to Lacan and 
Althusser. There are also the Cultural Studies people, who seek a 
politically engaged media analysis that unabashedly rejects the links 
with the avant-garde that were so central to much in Screen of the 1970s. 
(For these theorists, television, brimming with emergent readings and 
glimpses of Utopia, is plenty avant-garde enough.) This is not to men- 
tion the Deconstructionists, who threaten to turn theory utterly relativ- 
ist and dilettantish. In sum, SLAB (Saussure-Lacan-Althusser-Barthes) 
theory no longer holds undisputed sway. 

Perhaps this is why King spends so much time fretting over our work. 
It might somehow help sustain a comatose paradigm, with those pla- 
toons of theses taking shifts running the respirator. Of course. King’s 
concern about the impact of our work might be merely a culminating 
tranche. Screen’s recent issues on ‘Deconstructing Difference’ and 
Postmodernism may herald changes in Big Theory. 

In any event, the Wisconsin Project, although unmasked, remains at 
large. And we are used to surviving long winters. 


M.A. Course in the Social History of Art 

University Session 1988-89 

Applications are invited for a one calendar year full-time 
M.A. course in the Social History of Art. The course’s main 
component is a series of seminars and lectures on the prob- 
lems of theory and method in the construction of a social 
historical account of artistic production. Students also study 
a specific period or topic in art history in which they are 
encouraged to test out theoretical issues and procedures. In 
addition, students will produce a 10,000 word dissertation. 

Applicants should normally hold a good honours degree, 
and candidates are eligible to apply for B.A. Studentships. 
Further particulars and application forms are available from 
the M.A. Admissions Tutor, Department of Fine Art, Uni- 
versity of Leeds, Leeds LS2 9JT. 



’ The citation of \\ieke 
Bal comes from pl9. 
The reference to p25 
which I considered 
using as well but 
dropped in a rewriting, 
refers to actors who 
have no function in the 
Jabula, but arc 
nevertheless significant 
as sociological 

BEFORE MOVING TO the detail of my response, it is 
necessary, apparently, to comment on the term ‘the Wisconsin project’. 
Bordwell and Thompson - the former to the extent of gratuitously 
capitalising ‘project’ - are convinced that such a term is necessarily 
monolithic. As such it denies that they have other interests, not to men- 
tion the denial that there are other individuals active in cinema or film 
study at the University of Wisconsin. The term ‘project’ was deployed, 
in a loose Sartrean sense, to denote the fact that the various writers of the 
texts to be reviewed have as the object of their efforts the establishment 
of a distinctive theory of the classical Hollywood cinema, of cinematic 
narration and the activity of the spectator. Such a term does not entail 
absolute allegiance to all the propositions by all the participants nor 
does it deny differences of emphasis within it. What it does claim is that 
there is a sufficient level of convergence of concern to establish not iden- 
tity, but a ‘family’ resemblance. How one gets from this position to the 
claimed failure to appreciate that in other works the authors do other 
things or other researchers do other things, is not entirely clear. 

Turning to a direct consideration of the content of the replies, it is ap- 
parent that I cannot enter a response to every point raised. Some of the 
points are raised by all three authors and some seem of only peripheral 
or ad hominem significance. I shall of necessity be selective. But since 
that is precisely the way these authors responded to my original review, 
I feel that this is not an option I should deny myself. 


Before responding to David Bordwell directly, let me deal with one mat- 
ter on which we do agree. 

I apologise for the citation errors in the review and for the fact that 
checking procedures failed to iron them out. I can at least eat humble pie 
with a small sauce of consolation since no serious distortion is entailed, 
as Bordwell’s speculation that I might be bent on improvement (were 
that not impossible) suggests. Further, the difficulty of matters is only 
underscored by the fact that even Bordwell in his reply finds a passage 
(II, p 56) where there is none and says I leave out ‘the activities of the 
spectator’ when these are mentioned in the next paragraph (I, p 76). * 


This said, the chief difficulty I have in responding to his other points 
is that he has placed in my way a little ad hominem theatre - showing, as 
it were, a farce by Brian Rix in the tradition of Artaud - wherein I am 
strung according to his pleasure. To proceed I need to remove a few of 
the strings:. 

Bordwell evidently has problems with generalisations. Having gone 
on at length about the reifying horrors of the ‘Wisconsin project’, he is 
quite happy to posit a similar entity. I am, apparently, a quintessential 
moment of the Screen tradition. If it happens that I seem to hold views - 
my Bazin-flavoured view of referentiality - that do not fall readily into 
his idea of the tradition, then I am horribly confused. Likewise, since 
Screen is monolithically opposed to the kind of enquiry he sees himself 
undertaking, if I evidence any approval of the ‘project’ I am merely 
hedging my bets. Again, when I fail to give detailed attention to the 
parts of his work he thinks Screen ought to be interested in, it can only be 
that I have not read that part of his work. The blatant polemical value of 
these devices is plain, but their use has the consequence of leading Bord- 
well to misread my review at a number of points. To choose one exam- 
ple, he says that, following Screen's Lacanian orthodoxy, I cannot con- 
ceive of alternative psychoanalytical accounts of the cinema. Yet on the 
page he quotes I indicate, following Gertrud Koch, the possibility of a 
psychoanalysis of objects derived from the work of Sartre and Gaston 
Bachelard.^ Since Bordwell’s account leaves little space for either ap- 
proach, his opinion here is not entirely edifying. 

Next, and more perniciously, Bordwell is relentless in presenting me 
(and, by extension, any other reviewer) as absolutely required to offer 
counter-evidence or an alternative substantive theory. Since what I am 
actually doing is attempting to assess, the integrity of work that is 
(presumably) conclusive enough to set before the general reader in the 
area of film and media studies, such demands are beside the point. Such 
requirements, which in any case confuse a review with a refutation, 
would render the work unreviewable, short of total approval. This may 
be the point. 

Related to this self-serving ordinance is the expectation that any 
reviewer (or for that matter any reader) should have an intimate acquain- 
tance with the primary literature on which Bordwell builds his argu- 
ments. One could, indeed, go back to the primary literature, but then 
one would be reviewing Bordwell’s gloss of that literature and not his 
arguments in the texts under review. There is enough to do in respect of 
the latter - notwithstanding the question of whether Bordwell’s reading 
procedures are worthy of such attention. 

If these hypertrophied expectations were not enough, Bordwell 
motivates them within a melodramatic scenario in which it is taken as 
axiomatic that if I assess his arguments I am necessarily endorsing the 
positions he criticises. But this returns us to our point of entry into the 
little theatre of Professor Bordwell - the notion of a unitary Screen tradi- 
tion. Bordwell clearly needs this formation more than Screen does. 

But it is now time to get up out of my seat in that theatre. As for the 

^ Had Boidwdl been less 
obsessed with 
culminations he would 
have found a parallel 
reference in my ‘Screen 
Acting: Reflections on 
the Day*, Screen May- 
August 1986, vol 27 
nos 3-4, pp 134-139. 


^ Kristin Thompson, 
Eisen5tein*s *Ivan the 
Terrible*: a Neo- 
formalist AnalysiSy 
Princeton, Princeton 
University Press, 1981. 

drama, with its badgers, lumberjacks and rectifiers, I hope the readers 
have found it interesting and amusing. As for my opinion, I would be 
dishonest if I did not admit that Bordwell’s sense of humour reveals him 
as Marxist of the Zeppo tendency. 

Sec, for example, Alan 
Sociological Poetics and 
Aesthetic Theory, 
Alaemillan, 1986, p 68 



Turning to matters of interpretation and disagreement, given the format 
of Bordwell’s response I can do no other than respond in a point by 
point fashion. As far as possible, I will strive to make points of general 
relevance rather than engage in a personal dialogue, though, as readers 
will appreciate, this is no easy matter. 

Bordwell raises the question (and by implication suggests this is a pre- 
requisite for understanding his arguments) of my competence in For- 
malism. I am quite happy to admit that I am not expert in such matters 
and doubtless this defect led me to assume that Jameson’s and Bennett’s 
texts were more substantial than Bordwell, from the depth of his exper- 
tise, finds to be the case. To make matters worse, I held that these texts 
and, others besides, e.g., Thompson’s Ivan the Terrible^, would provide 
me with enough of a purchase on the basic terms to make a reasonable 
appraisal. There was, alas, a profundity I failed to tap. 

The clearest evidence of all this is my capsule statement, as Bordwell 
has it, of the Formalist position. But, as the review makes plain I am not 
offering a definition of the Formalist position, but defining what for- 
malism (small f) means in the context of these texts. The larger issue I 
am happy to leave to the experts, though even my level of acquaintance 
suggests that Bordwell’s use of Mukarovsky, for example, is selective, 
playing down the role of sociopolitical norms within the artwork. 
Granted that Bordwell prefers his own style of writing, it is nonetheless 
the case that my capsule statement, if not perfect, is reasonably adequate 
to its object, albeit not to the object Bordwell wishes to foist on it. Given 
this context, his objection to ‘a means of narration’ falls. The term ‘as a 
means of signification’ refers to the medium and not to a principle of for- 
malist analysis. There is a need, as indeed Janet Staiger points out, to 
distinguish between recognising the semiotic nature of a medium and 
doing an analysis based on semiotics. The reason for adding this 
qualifier here was to indicate that the writers give some recognition to 
matters of ideology and signification and do not assume, for example, 
Barry Salt’s ‘physicalist’ standpoint. 

I am quite happy to agree with Bordwell that the Formalists are con- 
cerned with describing a function not a medium. But I am sceptical 
about the rigour of such a distinction in the light of their analyses in 
CHC, for example, which strive to show how under definite socio- 
historical conditions specific functions become coterminous with 
cinema per se or ‘Hollywoodness’. I take it we can disagree about this. 
But I think there is a potential for confusing in their account this very 

Lastly, Bordwell finds it a serious fault that I use the term ‘Russian 
Formalist tradition’ without explicitly signalling the modifications in 
this tradition by the Prague School. This, Bordwell opines, is yet 
another symptom of my incomprehension. Rather, it is a symptom of 
my belief that the term tradition can be used as a term of location in a 
paradigm. I would use the term ‘Marxist tradition’ with equal equanimi- 
ty, even though it contains both Ralph Milliband and Nicos Poulantzas 
or, as Staiger suggests in her listing, Paul Hirst and Erik Olin Wright. 
These authors have very different positions but this does not refute a 
‘family’ resemblance. One might consider ‘the Screen tradition’, once 
again, in this light. ^ 

In a similar vein, Bordwell regards it as significant indicator of my in- 
ability to understand the basic arguments of his other key source, 
Cognitive Psychology, that I do not recognise that prototype and 
template schemata are both data processes. Other things being equal, 
/i7era//jy speaking he is correct. But other things are far from equal. In his 
account of narrative comprehension Bordwell emphasises that template 
schemata have the function of principles of higher level ordering or ‘fil- 
ing systems’ that integrate prototype schemata which are generically 
based. In addition, since he offers no other example, his account equates 
template schemata with the canonic story format which is supratextual 
or supra-generic - or if one prefers, in the West at least, culture-free. 
The question he evades is whether an account of the spectator’s activity 
can work at the level of abstraction from concrete readings that his 
prioritising emphasis on template schemata implies - his apparent ad- 
miration for the work of the Centre for Contemporary Cultural Studies 
is somewhat puzzling in this regard. 

The kind of competences I had in mind were those that are formed in 
the individual by virtue of his or her social location and patterns of 
cultural acquisition and consumption. (The work of Pierre Bourdieu is 
relevant here.) For me the problem is to show how a regularity such as 
the canonic story format emerges (if it does, since it might be an artefact 
of the level of analysis) out of a plurality of readings. Bordwell’s 
understanding of cognitive theory may be firm, but is his use of it? For 
example, even as a non-expert, I find his transportation of concepts from 
an account of memory to narrative comprehension warrants serious 
challenge. Again when he asserts that any other account of narrative 
comprehension will necessarily be driven to use the same concepts, he 
seems to equate a way of theorising cognitive activity with that activity 
itself. Since this error is held to be characteristic of cognitive theory in 
general, this is perhaps understandable.* As for environmentalism, 
Bordwell is clearly enjoying one of his jokes. Given his grasp of the field, 
he knows the term refers to the general position argued, for instance, by 
Gombrich in ‘Image and Code’^ or in the particular variant called an 
ecological approach by James Gibson®, that all that is necessary for 
perception is in the environment and not in the mind. Great fun! 

Turning to the question of ‘information theory’ Bordwell berates me 
for not understanding the difference between cognitive theory and in- 

® cf ‘Throughout, the 
theory elaborates 
concepts present in 
Russian Formalist 
work and the writings 
in that tradition by Jan 
Mukarovsky, Tzvetan 
Todorov, Gerard 
Genette, Meir 
Sternberg and Seymour 
Chatman’, David 
Bordwell, Narration in 
the Fiction Film, op cit, 
p xiii. 

* See Jeff Coulter, 
Rethinking Cognitive 
Theory, Basingstoke, 
Macmillan, 1983, 
chapter 1. 

^ EH Gombrich, 
‘Image and Code: 
Scope and Limits of 
Conventionalism in 

Representation’, The 
Image and the Eye, 
Oxford, Phaidon, 
1986, pp 278-297. 

® James J Gibson, The 
Ecological Approach to 
Visual Perception, 
Boston, Houghton 
Mifflin Co, 1979, 
chapter 16. 

* See Jeremy Campbell, 
Grammatical Ainfii 
Penguin, 1982.’ 

See David Bordwell, 
Narration in the 
Fiction Film, op cit, p 

* * See Edward Eranigan, 
' “Here is a Picture of 
No Revolver!” The 
Negation of Images 
and Methods for • 
Analysing the 
Structure of Pictorial 
Statements’, Wide 
Angle, vol 8 nos 3-4, 

formation theory, because the latter is a discrete theoretical tradition 
unrelated to the former. Only on a very strict and anachronistic defini- 
tion is this the case. The relevance span of information theory has mov- 
ed forward from Shannon’s terms of reference in 1948 and now can be 
held to cover a convergence of hitherto discrete fields - cybernetics, 
systems theory, information theory (in the strict sense), cognitive 
psychology, linguistics and more besides.’ Given this phenomenon of 
reflected meaning and the general configuration of their arguments- 
entailing systems thinking, e.g., the Classical style, information theory 
(Meyer'°), concepts derived from artificial intelligence which itself is 
developing as a match of cognitive theory and computer systems 
(\Vinograd“) - this use seems warranted. I am happy to admit some ter- 
minological hesitations - cognitive theory goes to the movies was too 
restrictive. Artificial intelligence goes to the movies might be construed 
as an insult. So I settled on information theory. I am quite happy to 
abandon the term. But not the substance of the point: that despite Bord- 
well’s agnosticism on ideology and the like, his theory of spectatorship 
entails a particular kind of subject. As for Staiger and Thompson, I take 
it that arguments can be supportive of a position without consciously 
adopting that position. 

Next, it seems trite, but I appear to need to remind Bordwell that I am 
offering a condensed appraisal of a body of writing that exceeds my 
review by a factor of at least 100. The only way to proceed in such a cir- 
cumstance, since after all is said and done I hope that readers will turn to 
the texts themselves, is to give my summary of the general tendency of 
the argument. Bordwell insists at great length that any general 
statements of tendency are empirical statements about the contents of 
the text(s). Thus, if I conclude that the articulation of narrative struc- 
ture and representation with narration is never subjected to a substantial 
analysis in his text, with the unacceptable consequence that narration is 
conflated with textuality, this is taken as an empirical statement about 
contents. Likewise, when I suggest that the theory of narration operates 
at a level of abstraction from the text so that the analysis of specific films 
seems to rely on a disparate and more conventional notion of genre and 
the like, this is taken as an empirical statement that the text contains no 
close analysis. If I conclude that narrative logic, which subsumes on this 
account space and time, is mainly rendered as causality and that as a con- 
sequence parallelism is assimilated to a causal function, this is inferred 
as a failure to mention a rigorously explored dimension of narrative. I 
would agree with this if Narration in the Fiction Film contained such an 
exploration, but on my reading it doesn’t. 

I am quite happy to have Bordwell point these and other emphases 
out, even if some seem trivial. But let us be clear: to mention something 
is not necessarily to articulate it into an account. Bordwell goes in for a 
lot of ‘technical’ coverage. It is the articulation of the concepts that I am 
seeking to identify. 



Having spent a brief while exposing some of the rhetorical underpin- 
nings of Bordwell’s response, I want now to address matters of straight- 
forward disagreement. 

Bordwell represents me as conflating a sign and its referent when, in 
fact, as the context makes clear (and anyone with a crib’s knowledge of 
semiotics would know), signs vary in respect of the degree to which the 
material of the signifier partakes of the signified. In other words, signs 
are more or less conventional, more or less motivated. Again, he seems 
to feel that indexicality is only a feature of documentary, when as a pro- 
ponent of formalism he should know that this sign function is only 
dominant in such a genre. Again, a painting is not an analogon in the 
same sense as a photograph (still or moving): the former may exhibit 
iconically-based resemblances to the referent, but indexicality only 
operates as a trace of the painter not the painted - though perhaps Bord- 
well has in mind dragging paint-daubed nudes over canvases. In the 
photographic image both relationships of reference are articulated. 
Generally speaking, as he amply demonstrates, Bordwell does not 
understand the distinction between an analogue and an analogon. Bar- 
thes’ use of the archaic term analogon was precisely to catch the dis- 
tinctiveness of photography. 

Bordwell points out correctly that I use the term ‘reading’ throughout 
when he claims to be doing an ‘analysis’. I clearly state that I think he is 
offering a reading despite what he claims. But it goes further than this: 
via some laboured wordplay he wants to suggest that the distinction bet- 
ween different kinds of reading and different kinds of competence is 
analogous to saying oranges and baseballs are the same thing. Of course, 
if you bite or whack the wrong one you know the difference. That’s em- 
piricism. Apart from the fact that it is not very sound logic to equate 
statements about complex entities with simple entities, Bordwell cannot 
even operate his own contrast which should be, strictly speaking {vide: 
his example of bats and robins sharing flight) baseballs and oranges 
grouped around some common property, e.g., sphericality. In other 
words, it is consistent for me to claim what he calls analysis is in fact a 
reading and to call for a contrast between his account and other 
readings. (On the question of description versus definition I will have 
more to say below.) Again, I don’t say some readers are smarter than 
others, I say smartness is an attribute which is only meaningful in a con- 
text which valorises a particular kind of reading. 

As for defining my terms, it should be obvious enough that I need in a 
review to assume a broad familiarity on the part of the readership with 
the terms I use. I do not think I am mistaken in this assumption. But for 
the sake of clarification let me say: discourse is used to identify the essen- 
tially dialogic nature of speech and writing and analogous forms of 
meaning production in the mechanical media. My Bakhtinian emphasis 
on the dialogic nature of discourse is meant to underscore the fact that 
any consideration of meaning cannot be successfully divorced from the 

MR Cohen and E 
Nagel, /Jn Introduction 
to Logic and Scientific 
Method, London, 
Routledge and Kegan 
Paul, 1972, pp 275 IT. 

social relations that prior to any item of discourse limit what is a valid 
form of expression and by extension who is a valid producer or con- 
sumer of that form and its associated content. Further, these latter rela- 
tionships (and others besides) need to be seen as articulations (relation- 
ships of power and knowledge) which determine what the meaning 
variation of a given text will be in a specifiable historical or social con- 
text. A text, as I see it, is always a problematic intersection of a number 
of discourses. A representation is a signifier which is tied, conditionally, 
to a relation of reference, which from the perspective of alternative 
discourses may appear reductive and arbitrary. I could go on, but I think 
readers will recognise the provenance of this sort of position, even if 
they do not agree with it. 

Rather let me say this: terms like subject, discourse, interpellation and 
the like are not free floating signifiers that can be shuffled as though they 
were merely formal equivalents. Bordwell wishes to establish that these 
terms are tautological, but all he actually establishes, apart from his 
dislike of the terms, is that if they are abstracted from their conditions of 
articulation they can be rendered as tautologies. In this he exhibits what 
might be called context blindness. The effect of this syndrome is rather 
analogous to evaluating the use of the term ‘elasticity’ in Economics by 
consulting underwear catalogues. 

Turning to the theory of spectator activity, it is apparently a telling 
flaw in my reading that I characterise his approach as hypothetico- 
deductive. On his account it is patently inductive, allhough it uses 
hypotheses. There is a need to distinguish between what a spectator 
does in ‘making sense’ of a particular film which may be presented as a 
wholly inductive process and the theorisation of the regularities (if there 
are such) that govern or condition such a process. But since the process 
of making inferences from the film unfolding is on Bordwell’s account 
conditioned by the operation of schemata, it follows that identifying 
what is a valid instance on which to base an inference is a process of 
deduction. In other words, inductive reasoning is, barring the possibili- 
ty of an enumeration of all the instances subsumable under a universal 
proposition, a form of deduction.'^ To generalise the point, I would 
argue that spectators do not arrive at the ‘text’ with an ‘open’ mind, but, 
on the contrary, with a range of expectations - some proper to the 
cinema per se and some proper to their position in extra-cinematic 
discursive practices - that condition inference making. A given film may 
confirm or upset such expectations - and I think it is a virtue of these 
writers’ work that it identifies some of the ways in which this can 
occur -but this does not mean that spectators are only operating 

I still remain puzzled why the spectator’s activity in historical mater- 
ialist film is said by Bordwell to be purely inductive. (Incidentally, I 
don’t say the Soviet films of the ’20s and ’30s are about ‘the dictatorship 
of the proletariat’. I say this is an example of the kind of didactic guide 
that this mode of narration might offer.) 

These considerations lead to the question of a realist reading. Bord- 


well represents my comments here as a denial of what is patently the 
case, that he offers the elements of a theory of how a film’s narrational 
strategy may lead to a spectator response that undervalues the film’s 
own discursive practices as opposed to other practices clustered around 
the point 0 / reception. Actually what I question is why he chooses to 
present thematic readings as ‘bad’ readings rather than as consistent 
readings given a specific level of competence in the spectator. Further- 
more, since I don’t accept that readings are exclusively a function of 
these aspects of textual organisation that Bordwell emphasises, it fol- 
lows that I find his treatment of thematic readings vestigial. He is cer- 
tainly entitled to claim that this is not what he is doing. But then he 
forfeits the right to pronounce on the validity of such readings. 

Turning to the central issue of style. I have already made it clear that 
my preference is for a conception of style that is linked to ideology. This 
is not Bordwell’s preference, rather he prefers to assume that style can 
be studied as an immanent phenomenon. Given this restriction, ably 
signalled by the citation from Beardsley, the pertinent issue becomes 
how to explain the mediation between the microcosmic (ideographic) 
sense of style and style as a macrocosmic phenomenon which carries the 
(nomological) sense of lawlike regularity. I don’t profess to be able to 
answer the problem of reconciling these approaches - though I think 
Hadjinicolaou'^ has interesting things to say in this regard. I do, 
however, recognise that it is a question pertinent to Bordwell’s enquiry 
which he may be reasonably expected to not merely identify but provi- 
sionally answer. After all, he claims to have demonstrated that a certain 
body of films can be characterised (despite individual differences) as 
belonging to a group style. If a particular film can be described as validly 
belonging to the classical cinema, then some of its immanent 
characteristics must pertain to this genus. In suggesting that the dif- 
ference between The Classical Hollywood Cinema and Narration in the 
Fiction Film is that of a conception of style, Bordwell throws into ques- 
tion the very concept of systematicity advanced in the first study. 

I read Bordwell as saying that the distinction between extrinsic and in- 
trinsic norms might offer a way of reconciling the two senses of style. 
But this is not so. Yet if it is true as he asserts {NiFF, p 150) that ‘a film 
may accede (differentially) to the reigning set of norms: nearly all 
Hollywood films do this’ and that such films may also set up intrinsic 
norms, ‘standards attained within the text itself (ibid), it is difficult, for 
me at any rate, to see how the two conceptions of style are not, in fact, in 
some sense interrelated in any mode of narration. 

If on the other hand, as he suggests, a deviation from an extrinsic norm 
can be considered in isolation from a deviation from an intrinsic norm, it 
is difficult to see how the concept of group style could survive the 
analysis of a specific film. In either case, I fail to see how the shift in the 
definition of style between The Classical Hollywood Cinema and Narra- 
tion in the Fiction Film can be regarded as irrelevant to their respective 

In the same connection. Bordwell berates me for makine too much of 

' ^ Nicos Hadjinicolaou, 
An History and Class 
Struggle, London, 
Pluto, 1978. 


Seymour Chatman, 
Review of Narration 
in the Fiction FUm, 
Wide Angle, vol 8 nos 
34 . 

the repositioning of syiizhet (plot) and narration between the same texts. 
But there is more at issue here than merely the spatial repositioning of 
terms. In The Classical Hollywood Cinema, narration is that aspect of the 
text in rofo- the syuzhei (plot) -that transmits story information, 
mobilising stylistic devices to attain this objective. The specific effec- 
tivity of narration rests on the manner in which ‘it’ transmits story 
information - with a variable degree of self-consciousness, knowledge- 
ability and communicativeness, A narrative film is argued to consist of 
three systems: narrative logic, the representation of time and of space. 
But the classical narrative film is marked by a specific articulation of 
these three systems, a dominant in which narrative causality is rendered 
through character and, pre-eminently, compositional motivation. It is 
this dominant, this tendency for the classical film to give an unequal 
weight to this specific kind of causality, that ‘provides’ the guidelines 
for the activity of narration. While Bordwell does not say explicitly that 
it is the syuzhet that ‘imposes’ these guidelines (he says ambiguously that 
the distinction syiizhet/fabula cuts across the three system distinction, 
CHC, p 13), it follows by definition that this is the case. At any rate, he 
does not explicitly identify any other immanent formative process of 
sufficient generality (‘the totality of formal and stylistic materials in the 
film’, NiFF, p 344) and the notion of a dominant warrants this conclu- 
sion. If the relationship between plot, narration and style in the classical 
film is hierarchical, so that style is subordinated to narration and narra- 
tion to syuzhet, it is not merely open to a ‘free’ spatial reallocation. If nar- 
ration now occupies the ‘slot’ once held by syuzhet, there is a (potential- 
ly) different kind of formative relationship. In other words, the slots are 
not equal in respect of their causal weight. 

Of the new relationship in Narration in the Fiction Film, Bordwell of- 
fers no account of what this shift entails, he merely describes it. Has nar- 
ration now assumed the role of inscribing the ‘dominant’ in classical 
film? If so what does this mean for Narration in the Fiction Film and its 

It is not my place to second guess Bordwell, but in my view the swit- 
ching between narration and syuzhet that founds the analysis in Narra- 
tion in the Fiction Film is spurious. This is because, I believe, that in this 
latter text, Bordwell effectively equates narration with the action of the 
syuzhet, which in classical narration, the crucial case, subtends style to 
‘its’ ordering. This means, in effect, that any film now entails a two-term 
system of interaction: syuzhet and style. Such an attenuation of the con- 
cept of narration is principally signalled by Bordwell’s formal defini- 
tion, which defines narration as a process constructed out of the interac- 
tion of syuzhet and style, rather than a third dynamic in the process of in- 
teraction. But it takes its theoretical warrant from the implicit decision 
to treat narration as a process without a narrating instance, e.g., an im- 
plied author’^ and, correspondingly, a concretely defined narratee - the 

As I see it, this elision is built on assuming the efiicacy of dimensions 
of narrative (that can be loosely defined as story constructing informa- 

tion) which underpin, but are never effectively articulated into, his ac- 
count. These dimensions I counterpose (somewhat repetitiously I now 
feel) to the higher level ordering of the syuzhet. The implied query - 
how does this lower level of determination which structures, in part, 
story information relate to Bordwell’s theory of narration - is brushed 
aside as an invitation to consider E M Forster. On the contrary, it is an 
invitation to consider a dimension that Forster’s original specification 
and Sternberg’s treatment point towards. 

One last matter needs to be clarified under the heading of style. Bord- 
well reads me as challenging his analysis of Bresson’s Pickpocket. But 
what I am actually doing is examining what happens to his definition of 
style in the course of his reading, not whether such a reading is intrin- 
sically valid. Given this restriction, I remain unconvinced that I need to 
give ground on the problems of parametric narration, even if I accept 
the corrections to my account Bordwell suggests. Moreover, I find 
myself puzzled by Bordwell’s indignation that I should quote Bresson 
back at him. Just as he says I should not reject a position I don’t like 
wholesale, e.g., formalism, so I take it that a dread of auteurism should 
not lead to the assumption that a director can never offer an adequate 
theorisation of his or her own practice. I detect a potential contradiction 
in Bordwell’s assertion that his conception of style has always entailed 
not merely camerawork, but other coding practices, such as set design, 
performance and the like, and his attempt to establish the terms of 
minimalist film style. Likewise, in his rejection of the view that enun- 
ciative markers might operate elsewhere in the diegesis, he shows that 
his view of what is pertinent to the representation of intentionality, 
which is a central, if problematic, ingredient of the concept style, is far 
from parsimonious. As with style, it is to be understood that his search 
for enunciative markers has already subsumed camerawork and the pro- 
filmic event, i.e., all the fabricative events that construct or re-present 
the profilmic. 

Given this generalising sweep it is hardly surprising that Bordwell can 
claim that further analysis is unlikely to prove fecund. In both cases, 
what is to be explained in the light of a definition is taken as already in- 
cluded in the definition. In the specific case of style, I hold the opinion 
that Bresson’s theorisation, centring on the image, offers a more consis- 
tent way of characterising the relationship between a minimalist com- 
positional strategy and textual richness. 

This brings me to the question of the fabula. Bordwell’s response is 
evasive. For many theorists, e.g., Culler'^ Chatman’*, the early 
Shklovsky'^, the distinction between fabula and syuzhet is a distinction 
between pre-existent story materials and the manner in which these pre- 
existent materials are ordered in the text. (A parallel use is found in film 
theory as in E Ann Kaplan’s distinction between discourse and 
diegesis.’®) These pre-existent story materials have both a ‘mental’ and 
material existence in that they are already pre-given as a stock of ideas 
and/or texts (or, indeed, performances). 

Bordwell’s use of the fabula! syuzhet couplet is quite different - a fact 

’* Jonathan Culler, The 
Pursuit of Signs f 
London, Routledge 
and Kegan Paul, 
1981, pp 169 ir. 

’* Seymour Chatman, 
Story and Discourse, 
Ithaca, New York, 
Cornell University 
Press, 1978. 

See L Matejka and K 
Pomorska (eds). 
Readings in Russian 
Poetics, Cambridge, 
Mass, MIT Press, 
1971, p 20. 

’® E Ann Kaplan, 
Women and Film: 
Both Sides of the 
Camera, London and 
New York, Methuen, 
1983, pp 18-19. 

** Michael Baxandall, 
Patierm cf Intention, 
New Haven and 
London, Yale 
University Press, 
1985, chapter 1. 

he notes but never actually justifies. For him, the term syuzhet begins to 
operate in a manner that conflates the ordering instance of the text and 
what is ordered (pre-existent ‘stuff) given that fabula is now rendered as 
a psychological entity. Let me note the obvious point, that if this con- 
trast were rendered phenomenologically the fabula in Bordwell’s sense 
would become a resource that the spectator brings to bear on a text, 
which in itself is likely to be determined by pre-existent materials and 
discourses - such is one of the senses of intertextuality. Such a general 
notion of what might be part of a ‘hermeneutic’ circle is not addressed in 
Bordwell’s account. 

Obviously if the fabula is exclusively an inferential entity, then it is 
quite inconsistent to suggest that it is a complex discursive formation 
which is brought to bear on the consumption of a given film. But I am 
not actually operating with Bordwell’s definition, but trying to prob- 
lematise it in the service of the definition I prefer. If I had to confine 
myself to Bordwell’s meaning then I would want to suggest that there is 
a need to make a distinction between inferences and percepts, with the lat- 
ter defined as already coded signals from the environment. Much of 
what I say about the analogon or about film as performance rests on this 
distinction. In connection with the materiality of the fabula, Bordwell 
feels I grievously misrepresent his sense by editing his statement about 
Jeff on screen in Rear Window. I have to admit that I felt that the phrase 
‘many other ways’ rendered the succeeding phrase ‘requiring no sight or 
sound of Jeff at all’ redundant. But, perhaps, the emphasis is worth ad- 
ding. It is, however, irrelevant to the point I am making and this em- 
pirical instance makes - sometimes elements of the fabula are on screen. 
When I read that, according to Bordwell, I claim their approach ‘aban- 
dons the question of meaning altogether’ and find that, unlike Thomp- 
son, he does not add ‘by means of rendering all content an equal expres- 
sion of form’, I don’t feel too penitent. 

This brings me to the question of definitions. Bordwell says that I do 
not understand the distinction between a definition and a description. I 
have, to an extent, to plead guilty because I hold to the view that a de- 
scription is a form of definition. To demonstrate this would require a 
long detour into formal logic which I confess myself not the best person 
to undertake. Perhaps I can satisfy (!) Bordwell by pointing him to the 
work of one of his heroes, Michael Baxandall ”, where this point is 
made more elegantly than I could make it. I feel reasonably confident, 
however, in saying the following: 

Bordwell tells me that his definitions are stipulative definitions, some- 
times called nominal definitions. Such definitions fix the meaning of a 
new term when it is introduced or, alternatively, seek to limit the mean- 
ing of a term which is well established, but woolly, to a particular mean- 
ing in a body of writing. Such definitions are neither true nor false, but 
rather propose that X will always mean Y in this context. In their worst 
form, and Bordwell will relish this Carrollian moment, stipulative defi- 
nitions mean that an author is saying, like Humpty Dumpty, that X 
means whatever he wants it to mean. Let me be gracious and say that 

Bordwell intends his definitions to be judged by their usefulness. I have 
given my opinion on that, which is clearly not the last word. But it 
would be remiss of me not to point out that much of Bordwell’s criticism 
of me rests on the defence of stipulative definitions. Nor do I think he 
helps his own case by insisting that his definitions are not bound by 
unique membership conditions or necessary or sufficient conditions. 
The first is plausible, the second is a hefty step towards Humpty 

Conversely, his definitions may be, as I originally believed, theoretical 
definitions. Such definitions, which are notoriously debatable, aim to 
formulate a theoretically adequate or scientifically useful characterisa- 
tion of the objects to which the term applies. Does Bordwell really want 
to claim that he is only offering stipulative definitions? I think he does 
his own work some disservice. 


In conclusion, the matter of empiricism: Bordwell and his co-writers 
grossly misrepresent my position here. When I say that the research 
undertaken would require a platoon of theses to evaluate it I mean just 
that. Many of the empirical claims in these works have in the short term 
to be taken at face value. My review offered itself as a test of the ade- 
quacy of the theory which I believe underwrites the empirical analysis. 
This is admittedly difficult, because even if the authors are more scru- 
pulous than other writers, they give very little detail of their underlying 
methodology. For example, as far as I can determine, the criterion of 
aesthetic pertinence governs the scrutiny of the films analysed, though 
the limitations of this criterion are not considered. Again, I am told that 
viewing procedures were uniform and what was enumerated, but not 
how the various elements from shot scale through to significant dialogue 
were weighted in importance or by what criteria they were rendered 
comparable. In short, how was uniformity achieved? In regard to these 
and other questions - how does the unbiased sample relate to the 
extended sample? - 1 am not inclined to be too sceptical. But given the 
fact that such questions are left open, Bordwell cannot honestly dis- 
qualify an evaluation of these particular texts that sets aside a detailed 
consideration of methodology if he does not provide such an account 

Bordwell sees a serious, not to mention crass, flaw in my suggestion 
that it might have been more effective to institute a random selection of 
films within a sampling frame that took as given, in the late silent 
period, the eight majors. In principle he feels this or any other frame 
would be absurd because ‘the point of the unbiased sample is exactly not 
to weight the selection by such a priori decisions’. He does not seem to 
have heard of stratified random sampling which is entirely legitimate 
when a given universe is prestructured: the issue is not whether the 
majors existed but rather whether the films they produced or distributed 

showed up certain constancies. If Bordwell’s point is taken to the letter 
it would follow that Barry Salt’s statistical style analysis^ which logs a 
device such as cutting rate without reference to its production - and 
indeed ideological - context, is more rigorous. Clearly, taking a random 
sample of films without respect to their studio location, sets aside diffi- 
cult questions of deciding which films are representative in the early 
silent period, given that the full institutionalisation of the classical sys- 
tem is achieved subsequently, but this merely argues for a different 
approach to ‘pre-studio’ films. 

I do not claim to know of other resources, though I suspect, as Kevin 
Brownlow has been demonstrating for years, that the list of survivors, 
and more important our notions of the completeness of the films that 
survive, is subject to a constant updating. I merely pointed out that had 
the procedure of stratified random sampling been adopted it would have 
strengthened our acceptance or rejection of the claim that the classical 
style transcended differences at the level of studio organisation. Since it 
is always a danger that the perceived uniformity is, in fact, an artefact of 
the analyst’s reading procedures (and this applies to us all) it is useful to 
make the test of the data as strong as possible. Without pronouncing on 
the issue of availability, it seems unlikely, in the post-silent period, that 
so few films produced or distributed by the majors are available from 
each to make random sampling within each slot impossible. If it turns 
out that what I suspect is not the case, then it may have been necessary 
to consider other ways of constituting the sample. Clearly the over- 
weighting of Warner Brothers is a problem. 

I can agree with Bordwell’s tautological assertion that the data he and 
his co-authors produce is better set to sustain the kind of generalisations 
they are seeking than analyses by other writers who use a single film. 
(But I have to leave open the question of whether these other authors are 
seeking to make the same kind of analysis.) I also agree that their 
methodology as presented sets new standards of rigour and evidence - in 
fact my review said that the high standard the authors set themselves 
means questions of data are worth raising. 

Lastly, since it is raised elsewhere, I need to point out the other empir- 
ical issue Bordwell raises: the Toland question. The point I am signal- 
ling here is this: does the conception of the classical cinema as a general 
process not conflict with the micronarrative of technical (and other) 
change in such a way as to install an overarching teleology? From the 
perspective of such a micronarrative, the notion of the classical cinema 
as remaining unchanged qa change, phis c’est la mime chose -looks 
like a disengagement between two distinct levels of analysis which the 
authors pursue at the same time. I am not claiming I could do better. But 
the characterisation of the classical cinema as an entity which is firm at 
its boundaries but, internally, is very flexible, can mean that change at 
the micronarrative level is by definition rendered as a sort of tinkering. 



Janet Staiger’s response, if read after Bordwell’s, is in many ways a 
model of restraint. Let me pass on the unhelpful opening in which she 
reads me as a kind of impossible theoretical object. I don’t say I have an 
allegiance to ‘deconstruction, Lacan feminism and postmodernism’. I 
say if one has an allegiance to any of these then the body of the work will 
look rather conservative, not to mention unremittingly hostile. Since I 
believe that a significant proportion of Screen readers will have an 
interest, however nuanced, in such conceptual frameworks, I want to 
clear the ground for a consideration of the texts in their own right. I 
clearly need to signal at various points how these texts provide problems 
for some of the concerns Screen represents. If I really saw myself as 
having an unruffled allegiance to all three projects I would be very con- 
fused indeed, since at various points they exhibit sharp paradigm con- 
flicts, not to mention differences in approach and method. I suppose this 
is why Staiger calls for ‘clinical’ evidence? Staiger’s ‘post-structuralist 
critique’ is for the most part an ironic device to say what David Bordwell 
says over and over: I am confused and unconstrained by scholarly con- 
ventions of consistency and logic in argument. (I may be so by default, 
but I am not so by intention.) I would prefer to respond to her in the con- 
text of a political economy of culture analysis, because this is where her 
criticisms are actually sited. 

The first point that Staiger advances is that I seriously misread her 
account of the role of economic and ideological/signifying practices 
within ‘the’ Hollywood mode of production. To this end, she cites a spe- 
cific instance which also combines a complaint about my (wilful) edit- 
ing. She suggests that I generalise from a specific example, which con- 
cerns the development of an extensive division of labour in editing 
which is not cost-effective, in the sense of lowering wage bills. The point 
she is making here is worth making: if one were to equate cost reduction 
with capitalist control in film-making or in labour processes as a whole 
this would be seriously misleading. I would take the view, hardly origi- 
nal to Staiger or me, that the costs of production in toto, let alone wage 
costs, will be inflated not merely because of the technical division of 
labour (which, in any case, is not a separable phenomenon from ques- 
tions of control), but because of the social division of labour in the work- 
place. In other words, the maintenance of capitalist control over a work 
process can be expensive. She can hardly complain if I use this as a gene- 
ral example of the thrust of her argument because that is what she is 

But she can complain if I use this general example in a way that mis- 
represents what she is trying to say. In the sentence I cite, as well as 
others, I read her as saying that ‘style’ is primary. In the following sen- 
tence which I regarded as redundant, she holds the opposite that econo- 
mic determinants are primary. Readers must weigh this up for them- 
selves but I read her succeeding statement not as reinstating the primacy 
of the economic, but rather sorting the relationship between the ‘stylis- 

tic’ and the ‘economic’ around necessary and sufTicient conditions. Style 
‘wins out’ because if the economic is a necessary condition (a condition 
without which there would be no production at all) the ‘stylistic’ is a suf- 
ficient condition which would account for the specificity of the 
Hollywood mode of production or, for that matter, a given film text. 
This is what I think she means in distinguishing between why labour 
was divided and how it was divided, the how being ‘influenced’ by ideo- 
logical/signifying practices. She further asserts that she does not equate 
the latter with ‘style’. I will leave it for the reader’s judgement to assess 
whether Staiger can validly claim that she is not operating a distinction 
between necessary and sufRcient conditions in the subsequent uncited 
sentence. I can only take it as an example of her enthusiasm to distance 
herself from my reading that she asserts the economic is primary in her 
account. Moreover, I can only take her word that her original formula- 
tion was so economistic, so committed to a crude either/or notion of 
causality that she now finds it embarrassing. So she should, since if her 
text is to be read in that way its function in the ‘Wisconsin project’ is 
even more problematic than I thought. I had assumed, not the least on 
the basis of her sources, e.g., John Ellis, not to mention Raymond Wil- 
liams, that she was attempting to argue for the salience of economic 
determinants in an account that preserved some determinative space for 
‘textual’ autonomy. 

Let me attack the substance of her point: I equate style with ideological/ 
signifying practices. In point offact, I think that it isr/iewho accomplishes 
this equation. If the compound term ideological/signifying practices 
were comprehensively articulated in her contribution to The Classical 
Hollywood Cinema I think she would have a point, but the fact is that she 
only identifies as pertinent practices which are typified by group style 
{CHC, p 88) or ‘the standard of the quality film’. Merely to hint at other 
discursive practices as being relevant without further elaboration either 
means that she thinks that these exhaust Ellis’ four conditions of film 
practice, or that for purposes of argument in The Classical Hollywood 
Cinema these are equated with stylistic factors. At any rate, her admis- 
sion that she could not give ideological/signifying practices the kind of 
weight she might have desired merely reflects my perception of its 
underdevelopment. My point about the fruitfulness of the conflict 
between technical and aesthetic conceptions of efficiency was aimed 
entirely to point up a missed opportunity for examining the contradic- 
tions of difierent discursive practices. The fact that Staiger can argue 
that ideological and signifying practices should not be equated and yet 
deploy the awkward oxymoron ‘ideological/signifying practices’ 
epitomises the underdevelopment of these practices in her account. 

The second way in which I allegedly misrepresent Staiger’s account is 
through the claim that she has written a formalist treatment of mode of 
production literature. The attribution formalist (as she astutely 
observes) is primarily pointing towards a sociological or historical for- 
malism. Namely, she applies the characteristics of the capitalist organis- 
ation of work in general, e.g., standardisation, differentiation and the 

like, unproblematically, to the sphere of aesthetic production. An 
account of the latter would not only have to look at general imperatives 
derived from the organisation of the labour process by agents of capital 
outside Hollywood and how these are realised within Hollywood, but 
also at the forms of worker resistance and informal organisations that 
mediate such imperatives in general and the specific mechanisms by 
which workers within Hollywood offer resistance to control. Further, I 
do not hold that the methods of dividing labour are immediately linked 
to the exploitation of labour in isolation from ideological/signifying 
practices. On the contrary such practices - which often relate to rela- 
tions of domination, rather than exploitation, among the workforce, e.g., 
male v female, skilled v unskilled, white v black, non-manual v manual, 
and so on - are the already pre-given mediating context of exploitation. 

Obviously Staiger’s account identifies specific mechanisms of control, 
e.g., the continuity script, which are pertinent to a more tightly context- 
ualised account. But it remains too generalised or abstract in its thrust. 
What is needed is not just an account of the institutionalisation of the 
continuity script as an example of standardisation, but how it actually 
functioned at MGM or Warner Brothers. Such an account should stipu- 
late, inter alia, the manner in which labour is subordinated to capital and 
the ideological and ‘signifying’ relationships through which this subor- 
dination is accomplished and reproduced. In other words, labour is not 
just exploited, it is given a place in an ideological schema which invar- 
iably mediates and seeks to legitimise exploitation, usually by dividing 
the workforce around industry specific categories like casual (costed) 
to picture) and permanent (costed to studio overhead), ‘creative’ and 
technical, as well as more general relations of domination mentioned 
above. It is precisely because Staiger advances a material definition of 
mode of production and, at the same time, assimilates ideological/signi- 
fying practices to those immediately pertinent to the classical style that I 
argue that a material definition leads to the primacy of style - not in 
general, but in her account. Naturally enough we will not agree on the 
extent to which something is effective or ‘assumed’ in an account if it is 

In a similar vein, Staiger feels I unfairly delete the ‘critical determin- 
ants’ in the theory of spectatorship of social class, aesthetic tradition and 
ideology which would show a representation of a history which is not 
formalist. I agree it would. But in the context of a forward glance at the 
theory of spectatorship actually offered, I think it is not merely that 
these determinants are too complex to be treated in full - they surely are 
- but that the theory offered is on a number of points incompatible with 
an account of their effectivity. I have only to offer Bordwell’s views on 
the relevance of ideology in this respect. As for appropriation, I beg par- 
don. I thought terms like ‘social class’, etc, were in the public domain 
long before any of us came to use them. Once again, I do not count a 
mention as an articulation. 

Lastly, in the contrast between ‘widely held’ and ‘hegemonically 
reproduced’, I am inviting the reader to consider whether the account of 

the widespread difTusion of classical norms details the contestation of 
their implementation - not as a once and for all installation of a self- 
regulating, open system - but as a recurrently challenged formation that 
negotiates its conditions of dominance. I am prepared to accept that 
Staiger wishes to offer such an account. But I am not convinced she does 


The next prong of Staiger’s rebuttal is the conception of class conflict in 
film history. The first point she raises concerns the mode of production. 
The thrust of my comments is clear enough] I think that her use of the 
term is overly productivist, concerned with work order and the like. I 
would hold that an adequate characterisation of ‘the’ Hollywood mode 
of production - if there is one mode - would entail an assessment of the 
inter-relationships between production, distribution and exhibition, 
which would detail: (a) the relationship between various capitals such as 
‘banks’ and ‘movies’ - relations of superordination or subordination - 
or even the same capital formation operating as producer or distributor 
and exhibitor, e.g., MGM and Loews; (b) relationships between the 
forms of labour in production, distribution and exhibition, which are 
not inconsequential for the way labour is subordinated at the point of 
production, e.g., the role of the International Alliance of Theatrical 
Stage Employees in Hollywood labour relations. 

Staiger wishes to flag these relations as pertinent to the ‘industry’ 
level, which may be contrasted to the more molecular conception of 
mode of production with which she operates. My problem with this is 
twofold: first, the character of the industry level and concretely, the 
place of a specific production company within it, cannot be separated 
from the mode of production in her sense. What is telling in her account 
is her failure to consider the dimension of formal and real subsumption, 
and particularly in the case of the latter the modalities of the control 
exercised within the formation of the labour process. On my reading of 
the studio period, it is precisely the relationships between sectors, and 
relatedly forms of finance, that gives the work organisation of this or 
that studio its particular form. This fact is all the more paradoxical given 
the specific systems of management which Staiger delineates so val- 
uably. (It is also in this connection that my remarks about the relevance 
of particular production personnel need to be considered.) Second, it is 
important to recognise that if Hollywood should be seen as a service 
industry, as Elsaesser suggests (and Staiger agrees), then one conse- 
quence of this undoubtedly correct position has been a dynamic priorit- 
isation of the ?neam of realisation over the tneans of production. 

To this one must add considerations of the nature of film as commod- 
ity, as a barer of multiple and multiplying use values; its intensive impli- 
cation (at least, as a mass commodity) in a short cycle of accumulation; 
its mode of consumption as a service which is only validated as a use 

value after it is consumed compared to the opposite relationship with, 
for instance, cars. These kinds of factors, which demand attention to the 
specificity of film as an aesthetic commodity, militate against any simple 
translation of categories derived chiefly from the study of the manufac- 
ture of highly functionalised commodities to film production. This is 
another example of formalism. 

As for the source of finance making a difference to the character of a 
film, I think this is an empirical possibility in so far as the source of 
finance, e.g., debt financing, introduces a specific dynamic to the realis- 
ation process or, in phenomenal terms, to the relationship between pro- 
duction costs and profitability as typified by amortisation schedules - an 
accountancy device for identifying successful or unsuccessful rates of 
earning. Such a dynamic affects in turn the nature of the production 
process and the extent of its production values. In this regard, I need 
only to point to Paul Kerr’s work on the 'B’ film noir}° In general terms 
what is an exploitation movie, if not, like pornography, a reductively 
functionalised text that raises finance on the basis of a quick rate of 

IfStaiger’s point is that such an account is not to hand, I agree. But to 
cite Robert Sklar’s Movie Made A/?:erica^' as foreclosing this enquiry is 
complacent to say the least. What Sklar is doing, as he does repeatedly in 
that text, is to mount a consumer sovereignty thesis so essential to his 
kind of cultural history. Since she finds Raymond Williams so illuminat- 
ing - as we all do - 1 suggest she considers the potential conflict between 
the market and ideological needs of capitalism outlined in his study 

Paul Kerr, ‘Out of 
What Past? Notes on 
the B Film Noir’, 
Screen Education nos 
32-33, Autumn 
1979/80, pp 45-65. 

Robert Sklar, Movie 
Made America, A 
Cultural History of 
A merican Movies, 
New York, Vintage 
Books, 1975. 

Raymond Williams, 
Culture, London, 
Fontana, 1981, 
especially chapter 4. 


Staiger’s final point is that I hold a workerist view of Hollywood labour 
relations, which equates employee status with class consciousness. Fur- 
ther, since I do this I ask of her what she, with her grasp of US labour 
history, does not wish to give: a class-centred history of Hollywood. It 
will doubtless surprise her that I agree, in broad terms, with the points 
she makes here. I do not equate employee resistance at the point of pro- 
duction with class consciousness in the traditional sense. Least of all in 
Hollywood, which is a classic case of a divided workforce in which rela- 
tions of domination overdetermine relations of exploitation and, 
because of the Guild structure and the operation of unions as labour 
brokers in a casualised labour market, contradictory class locations and 
intra-class conflicts abound. Again, I agree that ‘labourers’ ideologies of 
quality, particularly given the skill density of the labour process, are the 
site in which employer/employee conflicts manifest themselves. Where 
I seem to differ is that I see a need to relate these conflicts to the under- 
lying structure of exploitation. I would want to argue that it is the under- 
lying structure of exploitation that determines the kind of stakes that 
intra- and inter-union rivalry and employer/employee conflict seeks to 

appropriate and retain. I would expect Staiger’s account to explore in 
more detail the range of ideologies of quality or what is suppressed, if it 
is, by the norms of the classical style. Pursuing this line would - and this 
is the point - problematise the thesis of the fundamental unity of the 
classical system. 

I would also suggest that an adequate study of technical discourse or, 
indeed, the ideology of those who unproblematically endorse the official 
norms of this literature, needs to be set in the context of a substantive 
analysis of the labour process. To this line of analysis it is probably 
desirable to add a comparative analysis of the notions of quality uttered 
by industry leaders, and those inscribed in advertising, publicity and the 
like as discursive practices. I have a suspicion that Staiger has this in 
mind. I disagree that this can be assumed as ‘in’ The Classical Hollywood 

As to the claimed equation of employee resistance with class struggle, 
I think I have said enough on that score: my point is not that Staiger 
should offer a workerist version of Hollywood, but rather that she plays 
down the points of resistance to management objectives. Skilled workers 
are notoriously difficult to locate within a workerist schema or, indeed, 
within an account that argues that the ‘real’ issues are only wage-related. 
Very often, conceptions of quality seem paramount, with the question of 
whether these are ‘tactical’ or integral counters in the wage-effort bar- 
gain practically irrelevant. At the very least, there seems, on my reading 
of Hollywood labour relations, a strong presumption that the below the 
line costs of movie production would have been lower if standards of 
quality, which relate to the determination of factors such as wage scales 
(e.g., ‘golden hours’), had been as containable or as agreed as the notion 
of ‘the’ classical style suggests. 

Two more points seem worthy of mention: Staiger quotes me (I, p 80) 
as offering the opinion that growing demands for quality in production 
are not considered in the light of their potential to clash or conflict. She 
can point to some references in her text where this possibility is men- 
tioned. Let me admit that if I meant only a clash between, for example, 
clarity and stardom, then her mention of this potential challenges my 
observation. The real problem is that my original formulation was 
ambiguous, because it was the clash between different conceptions of 
clarity, verisimilitude and so on, that I had in mind as the general con- 
text of argument suggests. 

Next, the question of pseudo-closure, which she and Bordwell find a 
telling absence in my account. I am quite happy to have this pointed out. 
But I admit I took Bordwell literally when he wrote (in this case, with 
respect to authorial intrusion) ‘so powerful is the classical paradigm that 
it regulates what may violate it’ (CHC, p 81). More substantially, the 
notion of pseudo-closure needs to be considered in the light of the con- 
cept of ‘excess’. As I understand it, the term has three possible mean- 
ings: (1) the notion of a chronic destabilisation of the relationship 
between the signifier and the signified in favour of the play of signifiers 
down an ‘interminable’ syntagmatic chain; (2) the notion of figural 

excess, as proposed for example by Lyotard, which directs attention to 
the impact of the texture of the text on the perception of the text itself; (3) 
as the inherence in any given text of a subtext or suppressed Other, as in, 
for example, ‘colonial discourse’. Whatever the relationship between 
these various senses of excess, the point I wish to emphasise is that from 
the perspective of any of them, Bordwell’s notion of.closure, combined 
as it is anyway with the formal exclusion of a consideration of ‘excess’, is 
vestigial to say the least. 

Lastly, Staiger finds an internal contradiction in my remarks on the 
treatment of the star and the film noirand my comments on style as sup- 
pressing the constitutive role of meaning. The two statements are not at 
odds as she seems to think. Treating genre as a constitutive moment 
would not exclude the effectivity of discursive practices after the 
moment of constitution, but it would establish a different mode of analy- 
sis than that employed in The Classical Hollywood Cinema. 


Kristin Thompson has interesting things to say about the pre-history of 
publication of the texts under review. I nevertheless stand by my assess- 
ment of the relevant research facilities and funding available to US and 
UK scholars in this field, which is not intended to discredit British con- 
tributions. As a matter of fact, I don’t say that British Cinema History or 
British Cinema Now are speculative rather than conclusive. I say that the 
essay form is of necessity provisional. That I am not unusual in this 
opinion is clear from the following: 

The academic study of the cinema is still in its infancy in Britain. And as the 
opening chapter in this hook argues, what serious film history there is some- 
titnes suffers frotn a rather narrow and restricted focus. 

This book is tiot intended to be a definitive history of the British cinema, 
even if such a history were possible to produce. Rather it attempts to illumin- 
ate certain key developments and movements in the history of the British 
cinema through nineteen essays. . . . The selection has been determined by 
what seemed to us to be important or interesting; necessarily, it has also been 
determined by what is possible in an area where there is only a limited 
amount of research activity.^^ 

Thompson seems to operate, like Bordwell, with a rather agonistic view 
of academic enquiry. If I think ‘Grand Theory’ can be useful, according 
to her, I necessarily deprecate the essay form. What I think -as the 
review makes clear - is that the tension between these modes of enquiry 
is more desirable than either approach in isolation. For an either/or cast 
of mind this is doubtless profoundly unsatisfying. As for denigrating the 
work of ‘fine’ historians, which is certainly not my intention, I suppose 
this would depend on a careful appraisal of what she considers fine in 
historiography against what I consider fine. 

James Curran and 
Vincent Porter, 
Briiish Cinema 
History, London, 
Weidenfeld and 
Nicolson, 1983, p 1. 

Turning to the matter of appropriation, Thompson is as forthright as 
she thinks I am in claiming I wish to assimilate their work to a Marxist 
perspective. She is astute enough in recognising my interest in such a 
perspective. But I think the thrust of my review is to show how their 
account is not appropriable in the way she suggests. She complains that 
I never suggest that ‘future hjstorians’ should go out and confront the 
archives. This is because I take it for granted that an adequate refutation 
would require doing exactly that. In the short term, an evaluation of 
their work will have to rest on the criterion of internal consistency and 
an entirely different set of criteria that relate to its usefulness as a posi- 
tive heuristic programme. I take it those who pursue the latter will be 
prepared to some extent to live with the former. My closing remarks 
were pointing towards this possibility. As a matter of personal opinion, 
however, I doubt that a Marxist could consistently make piecemeal use 
of their work in this way. As to whether I am keen to tackle the labour of 
refutation, my answer is the same as before. Evidently, I am keen 
enough to turn from my own research to undertake an extended review. 

The point about aesthetic pertinence is simply this; that a methodol- 
ogy premised on such a criterion cannot enter a claim to have defined the 
classical style suigeneris^ unless it is assumed that style is only an aesthe- 
tic phenomenon. 

As regards the discussion of Citizen Kane, I am happy to agree with 
Thompson that if one analyses the same scene with different questions 
in mind one produces different answers. What else? I was using her 
example to emphasise how an analysis which seeks to establish the integ- 
rity of the continuity system (and by extension, the classical system) 
rests on a high level of abstraction from the text. Clearly, she agrees with 

Finally, Thompson has it that I think the criteria for successful innov- 
ation that she outlines are obvious and uninformative. That is not what I 
actually say. I doubt that such criteria, which are interesting in them- 
selves, can be cited as evidence of the pervasiveness of the classical style. 
In fact, this is to confuse the characteristics of the medium with a speci- 
fic function or functional complex. I think this is an unresolved tension 
in their account as I said earlier. But according to David Bordwell, 
distinguishing the medium and a function is intrinsic to a formalist 

Now it is time to make an exit. In view of the pronounced ad hominem 
tenor of my respondents’ remarks, I can allow myself a personal observa- 
tion. My objective in undertaking this review for Screen, was to bring to 
the readership’s attention a body of work that was in many ways outside 
of the mainstream of film and cultural studies in this country. In this 
regard, and in all modesty, I think I can claim to have achieved that. 



SCREEN: Reviewing the Woman’s Picture 

The recent publication of Christine Gledhill’s anthology of key essays on melodrama, 
Home Is Where The Heart Is (British Film Institute), and Mary Ann Doane’s The Desire 
to Desire (Indiana University Press), as well as a related season of events organised in 
1988 by the British Film Institute, offers an opportunity to take stock of the debates which 
have retrieved film melodrama from critical opprobrium and reconstructed it as an 
object of film theory. These have moved on from the early work on Minnelli and Sirk, in 
which melodrama was assessed for its capacity to contain or expose social conflict. In the 
wake of studies of the gaze, the theory of the genre has developed in relation to the issues 
of cinematic address and spectatorship and the question of their gendered character. 
This has motivated the turn to that subsection of melodrama, the woman’s film, because 
of its address to a female spectator and hence its representation of ‘woman’ in a space not 
governed by the dynamics of the male gaze and the fetishising of the woman’s image. 
This issue would address the adequacy of this newly ‘classic’ paradigm, focussing on 
concepts of the fetish (simulation) and the masquerade (dissimulation); cross*gender 
viewing and identification; the modem recycling of romantic conventions; and gay and 
lesbian appropriations of the genre. 

Submissions are requested by April 1, 1988. 

SCREEN: The Last Special Issue on ‘Race’ 

Questions of ‘race’, ethnicity and cultural differences have been at the forefront of recent 
developments in film practices and theoretical speculation. The emergence of new black 
independent films in Britain; the appeal of crossover and mainstream movies inflected 
by an interest in ‘otherness’ (e.g.. My Beautiful Laundrette) and ‘minority’ broadcasting, 
all point to important shifts and trends in contemporary cinema and television. But why 
has this arisen in the ’80s? The corresponding proliferation of theory on ‘colonial dis- 
course’ and debates on ‘Third Cinema’ have emphasised the limits of Euro/ethno- 
centrism in Film Studies. But what impact has this had outside academia on practices 
such as reviewing, criticism or film-making itself? 

We are therefore calling for papers which critically reassess historical and contem- 
porary configurations of ‘race’ in film culture. In particular, we encourage concrete 
analyses of theories and practices which address: the aesthetics of black/Third World 
film; images of ‘difference’ in Hollywood or the Euro-American avant-garde; different 
positions occupied by black/white spectators and audiences; intersections of sexual and 
‘racial’ difference and their implications for feminist and psychoanalytic perspectives; 
formations of productions, exhibition and reception in different national/intemational 
traditions. Called the ‘last special issue’ in the belief that value distinctions between 
‘centre’ and ‘margin’ are becoming more difficult to sustain, we especially welcome 
articles which address the salience of ‘race’ in the cultural crisis of modernity. 
Submissions are requested by June 1, 1988. 

Inquiries of submissions to: the Editor, Screen, 29 Old Compton Street, London WIV 
5PL, England.